corpus_id
stringlengths 7
12
| paper_id
stringlengths 9
16
| title
stringlengths 1
261
| abstract
stringlengths 70
4.02k
| source
stringclasses 1
value | bibtex
stringlengths 208
20.9k
| citation_key
stringlengths 6
100
|
|---|---|---|---|---|---|---|
arxiv-665301 | 2410.02884 | LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level Mathematical Reasoning | <|reference_start|>LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level Mathematical Reasoning: This paper presents an advanced mathematical problem-solving framework, LLaMA-Berry, for enhancing the mathematical reasoning ability of Large Language Models (LLMs). The framework combines Monte Carlo Tree Search (MCTS) with iterative Self-Refine to optimize the reasoning path and utilizes a pairwise reward model to evaluate different paths globally. By leveraging the self-critic and rewriting capabilities of LLMs, Self-Refine applied to MCTS (SR-MCTS) overcomes the inefficiencies and limitations of conventional step-wise and greedy search algorithms by fostering a more efficient exploration of solution spaces. Pairwise Preference Reward Model~(PPRM), inspired by Reinforcement Learning from Human Feedback (RLHF), is then used to model pairwise preferences between solutions, utilizing an Enhanced Borda Count (EBC) method to synthesize these preferences into a global ranking score to find better answers. This approach addresses the challenges of scoring variability and non-independent distributions in mathematical reasoning tasks. The framework has been tested on general and advanced benchmarks, showing superior performance in terms of search efficiency and problem-solving capability compared to existing methods like ToT and rStar, particularly in complex Olympiad-level benchmarks, including GPQA, AIME24 and AMC23.<|reference_end|> | arxiv | @article{zhang2024llama-berry:,
title={LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level
Mathematical Reasoning},
author={Di Zhang, Jianbo Wu, Jingdi Lei, Tong Che, Jiatong Li, Tong Xie,
Xiaoshui Huang, Shufei Zhang, Marco Pavone, Yuqiang Li, Wanli Ouyang,
Dongzhan Zhou},
journal={arXiv preprint arXiv:2410.02884},
year={2024},
archivePrefix={arXiv},
eprint={2410.02884},
primaryClass={cs.AI cs.CL}
} | zhang2024llama-berry: |
arxiv-665302 | 2410.02888 | Pseudo-Automation: How Labor-Offsetting Technologies Reconfigure Roles and Relationships in Frontline Retail Work | <|reference_start|>Pseudo-Automation: How Labor-Offsetting Technologies Reconfigure Roles and Relationships in Frontline Retail Work: Self-service machines are a form of pseudo-automation; rather than actually automate tasks, they offset them to unpaid customers. Typically implemented for customer convenience and to reduce labor costs, self-service is often criticized for worsening customer service and increasing loss and theft for retailers. Though millions of frontline service workers continue to interact with these technologies on a day-to-day basis, little is known about how these machines change the nature of frontline labor. Through interviews with current and former cashiers who work with self-checkout technologies, we investigate how technology that offsets labor from an employee to a customer can reconfigure frontline work. We find three changes to cashiering tasks as a result of self-checkout: (1) Working at self-checkout involved parallel demands from multiple customers, (2) self-checkout work was more problem-oriented (including monitoring and policing customers), and (3) traditional checkout began to become more demanding as easier transactions were filtered to self-checkout. As their interactions with customers became more focused on problem solving and rule enforcement, cashiers were often positioned as adversaries to customers at self-checkout. To cope with perceived adversarialism, cashiers engaged in a form of relational patchwork, using techniques like scapegoating the self-checkout machine and providing excessive customer service in order to maintain positive customer interactions in the face of potential conflict. Our findings highlight how even under pseudo-automation, workers must engage in relational work to manage and mend negative human-to-human interactions so that machines can be properly implemented in context.<|reference_end|> | arxiv | @article{moradi2024pseudo-automation:,
title={Pseudo-Automation: How Labor-Offsetting Technologies Reconfigure Roles
and Relationships in Frontline Retail Work},
author={Pegah Moradi, Karen Levy, Cristobal Cheyre},
journal={arXiv preprint arXiv:2410.02888},
year={2024},
archivePrefix={arXiv},
eprint={2410.02888},
primaryClass={cs.HC cs.CY}
} | moradi2024pseudo-automation: |
arxiv-665303 | 2410.02890 | Universally Optimal Watermarking Schemes for LLMs: from Theory to Practice | <|reference_start|>Universally Optimal Watermarking Schemes for LLMs: from Theory to Practice: Large Language Models (LLMs) boosts human efficiency but also poses misuse risks, with watermarking serving as a reliable method to differentiate AI-generated content from human-created text. In this work, we propose a novel theoretical framework for watermarking LLMs. Particularly, we jointly optimize both the watermarking scheme and detector to maximize detection performance, while controlling the worst-case Type-I error and distortion in the watermarked text. Within our framework, we characterize the universally minimum Type-II error, showing a fundamental trade-off between detection performance and distortion. More importantly, we identify the optimal type of detectors and watermarking schemes. Building upon our theoretical analysis, we introduce a practical, model-agnostic and computationally efficient token-level watermarking algorithm that invokes a surrogate model and the Gumbel-max trick. Empirical results on Llama-13B and Mistral-8$\times$7B demonstrate the effectiveness of our method. Furthermore, we also explore how robustness can be integrated into our theoretical framework, which provides a foundation for designing future watermarking systems with improved resilience to adversarial attacks.<|reference_end|> | arxiv | @article{he2024universally,
title={Universally Optimal Watermarking Schemes for LLMs: from Theory to
Practice},
author={Haiyun He, Yepeng Liu, Ziqiao Wang, Yongyi Mao, Yuheng Bu},
journal={arXiv preprint arXiv:2410.02890},
year={2024},
archivePrefix={arXiv},
eprint={2410.02890},
primaryClass={cs.CR cs.IT cs.LG math.IT}
} | he2024universally |
arxiv-665304 | 2410.02891 | Gait Optimization for Legged Systems Through Mixed Distribution Cross-Entropy Optimization | <|reference_start|>Gait Optimization for Legged Systems Through Mixed Distribution Cross-Entropy Optimization: Legged robotic systems can play an important role in real-world applications due to their superior load-bearing capabilities, enhanced autonomy, and effective navigation on uneven terrain. They offer an optimal trade-off between mobility and payload capacity, excelling in diverse environments while maintaining efficiency in transporting heavy loads. However, planning and optimizing gaits and gait sequences for these robots presents significant challenges due to the complexity of their dynamic motion and the numerous optimization variables involved. Traditional trajectory optimization methods address these challenges by formulating the problem as an optimization task, aiming to minimize cost functions, and to automatically discover contact sequences. Despite their structured approach, optimization-based methods face substantial difficulties, particularly because such formulations result in highly nonlinear and difficult to solve problems. To address these limitations, we propose CrEGOpt, a bi-level optimization method that combines traditional trajectory optimization with a black-box optimization scheme. CrEGOpt at the higher level employs the Mixed Distribution Cross-Entropy Method to optimize both the gait sequence and the phase durations, thus simplifying the lower level trajectory optimization problem. This approach allows for fast solutions of complex gait optimization problems. Extensive evaluation in simulated environments demonstrates that CrEGOpt can find solutions for biped, quadruped, and hexapod robots in under 10 seconds. This novel bi-level optimization scheme offers a promising direction for future research in automatic contact scheduling.<|reference_end|> | arxiv | @article{tsikelis2024gait,
title={Gait Optimization for Legged Systems Through Mixed Distribution
Cross-Entropy Optimization},
author={Ioannis Tsikelis, Konstantinos Chatzilygeroudis},
journal={arXiv preprint arXiv:2410.02891},
year={2024},
archivePrefix={arXiv},
eprint={2410.02891},
primaryClass={cs.RO cs.SY eess.SY}
} | tsikelis2024gait |
arxiv-665305 | 2410.02892 | The Role of Deductive and Inductive Reasoning in Large Language Models | <|reference_start|>The Role of Deductive and Inductive Reasoning in Large Language Models: Large Language Models (LLMs) have achieved substantial progress in artificial intelligence, particularly in reasoning tasks. However, their reliance on static prompt structures, coupled with limited dynamic reasoning capabilities, often constrains their adaptability to complex and evolving problem spaces. In this paper, we propose the Deductive and InDuctive(DID) method, which enhances LLM reasoning by dynamically integrating both deductive and inductive reasoning within the prompt construction process. Drawing inspiration from cognitive science, the DID approach mirrors human adaptive reasoning mechanisms, offering a flexible framework that allows the model to adjust its reasoning pathways based on task context and performance. We empirically validate the efficacy of DID on established datasets such as AIW and MR-GSM8K, as well as on our custom dataset, Holiday Puzzle, which presents tasks about different holiday date calculating challenges. By leveraging DID's hybrid prompt strategy, we demonstrate significant improvements in both solution accuracy and reasoning quality, achieved without imposing substantial computational overhead. Our findings suggest that DID provides a more robust and cognitively aligned framework for reasoning in LLMs, contributing to the development of advanced LLM-driven problem-solving strategies informed by cognitive science models.<|reference_end|> | arxiv | @article{cai2024the,
title={The Role of Deductive and Inductive Reasoning in Large Language Models},
author={Chengkun Cai, Xu Zhao, Haoliang Liu, Zhongyu Jiang, Tianfang Zhang,
Zongkai Wu, Jenq-Neng Hwang, Lei Li},
journal={arXiv preprint arXiv:2410.02892},
year={2024},
archivePrefix={arXiv},
eprint={2410.02892},
primaryClass={cs.AI cs.CL cs.LG}
} | cai2024the |
arxiv-665306 | 2410.02894 | Task-Decoupled Image Inpainting Framework for Class-specific Object Remover | <|reference_start|>Task-Decoupled Image Inpainting Framework for Class-specific Object Remover: Object removal refers to the process of erasing designated objects from an image while preserving the overall appearance. Existing works on object removal erase removal targets using image inpainting networks. However, image inpainting networks often generate unsatisfactory removal results. In this work, we find that the current training approach which encourages a single image inpainting model to handle both object removal and restoration tasks is one of the reasons behind such unsatisfactory result. Based on this finding, we propose a task-decoupled image inpainting framework which generates two separate inpainting models: an object restorer for object restoration tasks and an object remover for object removal tasks. We train the object restorer with the masks that partially cover the removal targets. Then, the proposed framework makes an object restorer to generate a guidance for training the object remover. Using the proposed framework, we obtain a class-specific object remover which focuses on removing objects of a target class, aiming to better erase target class objects than general object removers. We also introduce a data curation method that encompasses the image selection and mask generation approaches used to produce training data for the proposed class-specific object remover. Using the proposed curation method, we can simulate the scenarios where an object remover is trained on the data with object removal ground truth images. Experiments on multiple datasets show that the proposed class-specific object remover can better remove target class objects than object removers based on image inpainting networks.<|reference_end|> | arxiv | @article{oh2024task-decoupled,
title={Task-Decoupled Image Inpainting Framework for Class-specific Object
Remover},
author={Changsuk Oh, H. Jin Kim},
journal={arXiv preprint arXiv:2410.02894},
year={2024},
archivePrefix={arXiv},
eprint={2410.02894},
primaryClass={cs.CV}
} | oh2024task-decoupled |
arxiv-665307 | 2410.02895 | Approximation Schemes for POMPDs with Continuous Spaces and Their Near Optimality | <|reference_start|>Approximation Schemes for POMPDs with Continuous Spaces and Their Near Optimality: We study an approximation method for partially observed Markov decision processes (POMDPs) with continuous spaces. Belief MDP reduction, which has been the standard approach to study POMDPs requires rigorous approximation methods for practical applications, due to the state space being lifted to the space of probability measures. Generalizing recent work, in this paper we present rigorous approximation methods via discretizing the observation space and constructing a fully observed finite MDP model using a finite length history of the discrete observations and control actions. We show that the resulting policy is near-optimal under some regularity assumptions on the channel, and under certain controlled filter stability requirements for the hidden state process. Furthermore, by quantizing the measurements, we are able to utilize refined filter stability conditions. We also provide a Q learning algorithm that uses a finite memory of discretized information variables, and prove its convergence to the optimality equation of the finite fully observed MDP constructed using the approximation method.<|reference_end|> | arxiv | @article{kara2024approximation,
title={Approximation Schemes for POMPDs with Continuous Spaces and Their Near
Optimality},
author={Ali Devran Kara, Erhan Bayraktar, Serdar Yuksel},
journal={arXiv preprint arXiv:2410.02895},
year={2024},
archivePrefix={arXiv},
eprint={2410.02895},
primaryClass={math.OC cs.SY eess.SY}
} | kara2024approximation |
arxiv-665308 | 2410.02897 | Cognitive Biases in Large Language Models for News Recommendation | <|reference_start|>Cognitive Biases in Large Language Models for News Recommendation: Despite large language models (LLMs) increasingly becoming important components of news recommender systems, employing LLMs in such systems introduces new risks, such as the influence of cognitive biases in LLMs. Cognitive biases refer to systematic patterns of deviation from norms or rationality in the judgment process, which can result in inaccurate outputs from LLMs, thus threatening the reliability of news recommender systems. Specifically, LLM-based news recommender systems affected by cognitive biases could lead to the propagation of misinformation, reinforcement of stereotypes, and the formation of echo chambers. In this paper, we explore the potential impact of multiple cognitive biases on LLM-based news recommender systems, including anchoring bias, framing bias, status quo bias and group attribution bias. Furthermore, to facilitate future research at improving the reliability of LLM-based news recommender systems, we discuss strategies to mitigate these biases through data augmentation, prompt engineering and learning algorithms aspects.<|reference_end|> | arxiv | @article{lyu2024cognitive,
title={Cognitive Biases in Large Language Models for News Recommendation},
author={Yougang Lyu, Xiaoyu Zhang, Zhaochun Ren, Maarten de Rijke},
journal={arXiv preprint arXiv:2410.02897},
year={2024},
archivePrefix={arXiv},
eprint={2410.02897},
primaryClass={cs.IR cs.AI cs.CL}
} | lyu2024cognitive |
arxiv-665309 | 2410.02898 | Solving Reach-Avoid-Stay Problems Using Deep Deterministic Policy Gradients | <|reference_start|>Solving Reach-Avoid-Stay Problems Using Deep Deterministic Policy Gradients: Reach-Avoid-Stay (RAS) optimal control enables systems such as robots and air taxis to reach their targets, avoid obstacles, and stay near the target. However, current methods for RAS often struggle with handling complex, dynamic environments and scaling to high-dimensional systems. While reinforcement learning (RL)-based reachability analysis addresses these challenges, it has yet to tackle the RAS problem. In this paper, we propose a two-step deep deterministic policy gradient (DDPG) method to extend RL-based reachability method to solve RAS problems. First, we train a function that characterizes the maximal robust control invariant set within the target set, where the system can safely stay, along with its corresponding policy. Second, we train a function that defines the set of states capable of safely reaching the robust control invariant set, along with its corresponding policy. We prove that this method results in the maximal robust RAS set in the absence of training errors and demonstrate that it enables RAS in complex environments, scales to high-dimensional systems, and achieves higher success rates for the RAS task compared to previous methods, validated through one simulation and two high-dimensional experiments.<|reference_end|> | arxiv | @article{chenevert2024solving,
title={Solving Reach-Avoid-Stay Problems Using Deep Deterministic Policy
Gradients},
author={Gabriel Chenevert, Jingqi Li, Achyuta kannan, Sangjae Bae, Donggun Lee},
journal={arXiv preprint arXiv:2410.02898},
year={2024},
archivePrefix={arXiv},
eprint={2410.02898},
primaryClass={eess.SY cs.LG cs.RO cs.SY}
} | chenevert2024solving |
arxiv-665310 | 2410.02899 | FactCheckmate: Preemptively Detecting and Mitigating Hallucinations in LMs | <|reference_start|>FactCheckmate: Preemptively Detecting and Mitigating Hallucinations in LMs: Language models (LMs) hallucinate. We inquire: Can we detect and mitigate hallucinations before they happen? This work answers this research question in the positive, by showing that the internal representations of LMs provide rich signals that can be used for this purpose. We introduce FactCheckMate, which preemptively detects hallucinations by learning a classifier that predicts whether the LM will hallucinate, based on the model's hidden states produced over the inputs, before decoding begins. If a hallucination is detected, FactCheckMate then intervenes, by adjusting the LM's hidden states such that the model will produce more factual outputs. FactCheckMate provides fresh insights that the inner workings of LMs can be revealed by their hidden states. Practically, both the detection and mitigation models in FactCheckMate are lightweight, adding little inference overhead; FactCheckMate proves a more efficient approach for mitigating hallucinations compared to many post-hoc alternatives. We evaluate FactCheckMate over LMs of different scales and model families (including Llama, Mistral, and Gemma), across a variety of QA datasets from different domains. Our results demonstrate the effectiveness of leveraging internal representations for early hallucination detection and mitigation, achieving over 70% preemptive detection accuracy. On average, outputs generated by LMs with intervention are 34.4% more factual compared to those without intervention. The average overhead difference in the inference time introduced by FactCheckMate is around 3.16 seconds.<|reference_end|> | arxiv | @article{alnuhait2024factcheckmate:,
title={FactCheckmate: Preemptively Detecting and Mitigating Hallucinations in
LMs},
author={Deema Alnuhait, Neeraja Kirtane, Muhammad Khalifa, Hao Peng},
journal={arXiv preprint arXiv:2410.02899},
year={2024},
archivePrefix={arXiv},
eprint={2410.02899},
primaryClass={cs.CL}
} | alnuhait2024factcheckmate: |
arxiv-665311 | 2410.02901 | GTQCP: Greedy Topology-Aware Quantum Circuit Partitioning | <|reference_start|>GTQCP: Greedy Topology-Aware Quantum Circuit Partitioning: We propose Greedy Topology-Aware Quantum Circuit Partitioning (GTQCP), a novel quantum gate circuit partitioning method which partitions circuits by applying a greedy heuristic to the qubit dependency graph of the circuit. GTQCP is compared against three other gate partitioning methods, two of which (QuickPartitioner and ScanPartitioner) are part of the Berkley Quantum Synthesis Toolkit. GTQCP is shown to have 18% run time improvement ratio over the fastest approach (QuickPartitioner), and a 96% improvement over the highest quality approach (ScanPartitioner). The algorithm also demonstrates nearly identical result quality (number of partitions) compared with ScanPartitioner, and a 38% quality improvement over QuickPartitioner.<|reference_end|> | arxiv | @article{clark2024gtqcp:,
title={GTQCP: Greedy Topology-Aware Quantum Circuit Partitioning},
author={Joseph Clark, Travis S. Humble, and Himanshu Thapliyal},
journal={2023 IEEE International Conference on Quantum Computing and
Engineering (QCE), 2023, pp. 739-744},
year={2024},
doi={10.1109/QCE57702.2023.00089},
archivePrefix={arXiv},
eprint={2410.02901},
primaryClass={quant-ph cs.ET}
} | clark2024gtqcp: |
arxiv-665312 | 2410.02902 | Better Instruction-Following Through Minimum Bayes Risk | <|reference_start|>Better Instruction-Following Through Minimum Bayes Risk: General-purpose LLM judges capable of human-level evaluation provide not only a scalable and accurate way of evaluating instruction-following LLMs but also new avenues for supervising and improving their performance. One promising way of leveraging LLM judges for supervision is through Minimum Bayes Risk (MBR) decoding, which uses a reference-based evaluator to select a high-quality output from amongst a set of candidate outputs. In the first part of this work, we explore using MBR decoding as a method for improving the test-time performance of instruction-following LLMs. We find that MBR decoding with reference-based LLM judges substantially improves over greedy decoding, best-of-N decoding with reference-free judges and MBR decoding with lexical and embedding-based metrics on AlpacaEval and MT-Bench. These gains are consistent across LLMs with up to 70B parameters, demonstrating that smaller LLM judges can be used to supervise much larger LLMs. Then, seeking to retain the improvements from MBR decoding while mitigating additional test-time costs, we explore iterative self-training on MBR-decoded outputs. We find that self-training using Direct Preference Optimisation leads to significant performance gains, such that the self-trained models with greedy decoding generally match and sometimes exceed the performance of their base models with MBR decoding.<|reference_end|> | arxiv | @article{wu2024better,
title={Better Instruction-Following Through Minimum Bayes Risk},
author={Ian Wu, Patrick Fernandes, Amanda Bertsch, Seungone Kim, Sina Pakazad,
Graham Neubig},
journal={arXiv preprint arXiv:2410.02902},
year={2024},
archivePrefix={arXiv},
eprint={2410.02902},
primaryClass={cs.CL cs.AI}
} | wu2024better |
arxiv-665313 | 2410.02903 | Dissipative Avoidance Feedback for Reactive Navigation Under Second-Order Dynamics | <|reference_start|>Dissipative Avoidance Feedback for Reactive Navigation Under Second-Order Dynamics: This paper introduces DAF (Dissipative Avoidance Feedback), a novel approach for autonomous robot navigation in unknown, obstacle-filled environments with second-order dynamics. Unlike traditional APF (Artificial Potential Field) methods, which rely on repulsive forces based solely on position, DAF employs a dissipative feedback mechanism that adjusts the robot's motion in response to both its position and velocity, ensuring smoother, more natural obstacle avoidance. The proposed continuously differentiable controller solves the motion-to-goal problem while guaranteeing collision-free navigation by considering the robot's state and local obstacle distance information. We show that the controller guarantees safe navigation in generic $n$-dimensional environments and that all undesired $\omega$-limit points are unstable under certain \textit{controlled} curvature conditions. Designed for real-time implementation, DAF requires only locally measured data from limited-range sensors (e.g., LiDAR, depth cameras), making it particularly effective for robots navigating unknown workspaces.<|reference_end|> | arxiv | @article{smaili2024dissipative,
title={Dissipative Avoidance Feedback for Reactive Navigation Under
Second-Order Dynamics},
author={Lyes Smaili, Zhiqi Tang, Soulaimane Berkane and Tarek Hamel},
journal={arXiv preprint arXiv:2410.02903},
year={2024},
archivePrefix={arXiv},
eprint={2410.02903},
primaryClass={eess.SY cs.SY}
} | smaili2024dissipative |
arxiv-665314 | 2410.02904 | Convergence Guarantees for Neural Network-Based Hamilton-Jacobi Reachability | <|reference_start|>Convergence Guarantees for Neural Network-Based Hamilton-Jacobi Reachability: We provide a novel uniform convergence guarantee for DeepReach, a deep learning-based method for solving Hamilton-Jacobi-Isaacs (HJI) equations associated with reachability analysis. Specifically, we show that the DeepReach algorithm, as introduced by Bansal et al. in their eponymous paper from 2020, is stable in the sense that if the loss functional for the algorithm converges to zero, then the resulting neural network approximation converges uniformly to the classical solution of the HJI equation, assuming that a classical solution exists. We also provide numerical tests of the algorithm, replicating the experiments provided in the original DeepReach paper and empirically examining the impact that training with a supremum norm loss metric has on approximation error.<|reference_end|> | arxiv | @article{hofgard2024convergence,
title={Convergence Guarantees for Neural Network-Based Hamilton-Jacobi
Reachability},
author={William Hofgard},
journal={arXiv preprint arXiv:2410.02904},
year={2024},
archivePrefix={arXiv},
eprint={2410.02904},
primaryClass={math.OC cs.NA math.NA stat.ML}
} | hofgard2024convergence |
arxiv-665315 | 2410.02907 | NNetscape Navigator: Complex Demonstrations for Web Agents Without a Demonstrator | <|reference_start|>NNetscape Navigator: Complex Demonstrations for Web Agents Without a Demonstrator: We introduce NNetscape Navigator (NNetnav), a method for training web agents entirely through synthetic demonstrations. These demonstrations are collected by first interacting with a browser to generate trajectory rollouts, which are then retroactively labeled into instructions using a language model. Most work on training browser agents has relied on expensive human supervision, and the limited previous work on such interaction-first synthetic data techniques has failed to provide effective search through the exponential space of exploration. In contrast, NNetnav exploits the hierarchical structure of language instructions to make this search more tractable: complex instructions are typically decomposable into simpler subtasks, allowing NNetnav to automatically prune interaction episodes when an intermediate trajectory cannot be annotated with a meaningful sub-task. We use NNetnav demonstrations from a language model for supervised fine-tuning of a smaller language model policy, and find improvements of 6 points on WebArena and over 20 points on MiniWoB++, two popular environments for web-agents. Notably, on WebArena, we observe that language model policies can be further enhanced when fine-tuned with NNetnav demonstrations derived from the same language model. Finally, we collect and release a dataset of over 6k NNetnav demonstrations on WebArena, spanning a diverse and complex set of instructions.<|reference_end|> | arxiv | @article{murty2024nnetscape,
title={NNetscape Navigator: Complex Demonstrations for Web Agents Without a
Demonstrator},
author={Shikhar Murty and Dzmitry Bahdanau and Christopher D. Manning},
journal={arXiv preprint arXiv:2410.02907},
year={2024},
archivePrefix={arXiv},
eprint={2410.02907},
primaryClass={cs.CL}
} | murty2024nnetscape |
arxiv-665316 | 2410.02912 | Fine-Tuning Language Models with Differential Privacy through Adaptive Noise Allocation | <|reference_start|>Fine-Tuning Language Models with Differential Privacy through Adaptive Noise Allocation: Language models are capable of memorizing detailed patterns and information, leading to a double-edged effect: they achieve impressive modeling performance on downstream tasks with the stored knowledge but also raise significant privacy concerns. Traditional differential privacy based training approaches offer robust safeguards by employing a uniform noise distribution across all parameters. However, this overlooks the distinct sensitivities and contributions of individual parameters in privacy protection and often results in suboptimal models. To address these limitations, we propose ANADP, a novel algorithm that adaptively allocates additive noise based on the importance of model parameters. We demonstrate that ANADP narrows the performance gap between regular fine-tuning and traditional DP fine-tuning on a series of datasets while maintaining the required privacy constraints.<|reference_end|> | arxiv | @article{li2024fine-tuning,
title={Fine-Tuning Language Models with Differential Privacy through Adaptive
Noise Allocation},
author={Xianzhi Li, Ran Zmigrod, Zhiqiang Ma, Xiaomo Liu, Xiaodan Zhu},
journal={arXiv preprint arXiv:2410.02912},
year={2024},
archivePrefix={arXiv},
eprint={2410.02912},
primaryClass={cs.AI cs.CL cs.CR cs.LG}
} | li2024fine-tuning |
arxiv-665317 | 2410.02914 | Streamlining Conformal Information Retrieval via Score Refinement | <|reference_start|>Streamlining Conformal Information Retrieval via Score Refinement: Information retrieval (IR) methods, like retrieval augmented generation, are fundamental to modern applications but often lack statistical guarantees. Conformal prediction addresses this by retrieving sets guaranteed to include relevant information, yet existing approaches produce large-sized sets, incurring high computational costs and slow response times. In this work, we introduce a score refinement method that applies a simple monotone transformation to retrieval scores, leading to significantly smaller conformal sets while maintaining their statistical guarantees. Experiments on various BEIR benchmarks validate the effectiveness of our approach in producing compact sets containing relevant information.<|reference_end|> | arxiv | @article{intrator2024streamlining,
title={Streamlining Conformal Information Retrieval via Score Refinement},
author={Yotam Intrator, Ori Kelner, Regev Cohen, Roman Goldenberg, Ehud
Rivlin, Daniel Freedman},
journal={arXiv preprint arXiv:2410.02914},
year={2024},
archivePrefix={arXiv},
eprint={2410.02914},
primaryClass={cs.IR cs.AI cs.LG}
} | intrator2024streamlining |
arxiv-665318 | 2410.02915 | Does the Order of Fine-tuning Matter and Why? | <|reference_start|>Does the Order of Fine-tuning Matter and Why?: To improve the performance on a target task, researchers have fine-tuned language models with an intermediate task before the target task of interest. However, previous works have focused on the pre-trained language models and downstream tasks in Natural Language Processing (NLP) and considered only one intermediate task. The effect of fine-tuning multiple intermediate tasks and their ordering on target task performance has not been fully explored in Software Engineering. In this study, we perform the first empirical study on analyzing the impact of task ordering on target task performance. Experimental results show that there is an impact of task ordering on target task performance by up to 6% of performance gain and up to 4% of performance loss. To explain such an impact, we consider a variety of potential factors, including the characteristics of dataset (syntactic similarity and semantic similarity analysis, dataset size), model (probing task and attention analysis), and task (task affinity analysis). Our study provides Software Engineering researchers and practitioners with insights into the effect of task orderings and how to select the one that is cost-effective while achieving the best performance gain.<|reference_end|> | arxiv | @article{chen2024does,
title={Does the Order of Fine-tuning Matter and Why?},
author={Qihong Chen, Jiawei Li, Hyunjae Suh, Lianghao Jiang, Zheng Zhou,
Jingze Chen, Jiri Gesi, Iftekhar Ahmed},
journal={arXiv preprint arXiv:2410.02915},
year={2024},
archivePrefix={arXiv},
eprint={2410.02915},
primaryClass={cs.SE}
} | chen2024does |
arxiv-665319 | 2410.02916 | Safeguard is a Double-edged Sword: Denial-of-service Attack on Large Language Models | <|reference_start|>Safeguard is a Double-edged Sword: Denial-of-service Attack on Large Language Models: Safety is a paramount concern of large language models (LLMs) in their open deployment. To this end, safeguard methods aim to enforce the ethical and responsible use of LLMs through safety alignment or guardrail mechanisms. However, we found that the malicious attackers could exploit false positives of safeguards, i.e., fooling the safeguard model to block safe content mistakenly, leading to a new denial-of-service (DoS) attack on LLMs. Specifically, by software or phishing attacks on user client software, attackers insert a short, seemingly innocuous adversarial prompt into to user prompt templates in configuration files; thus, this prompt appears in final user requests without visibility in the user interface and is not trivial to identify. By designing an optimization process that utilizes gradient and attention information, our attack can automatically generate seemingly safe adversarial prompts, approximately only 30 characters long, that universally block over 97\% of user requests on Llama Guard 3. The attack presents a new dimension of evaluating LLM safeguards focusing on false positives, fundamentally different from the classic jailbreak.<|reference_end|> | arxiv | @article{zhang2024safeguard,
title={Safeguard is a Double-edged Sword: Denial-of-service Attack on Large
Language Models},
author={Qingzhao Zhang, Ziyang Xiong, Z. Morley Mao},
journal={arXiv preprint arXiv:2410.02916},
year={2024},
archivePrefix={arXiv},
eprint={2410.02916},
primaryClass={cs.CR cs.AI}
} | zhang2024safeguard |
arxiv-665320 | 2410.02917 | Deep image-based Adaptive BRDF Measure | <|reference_start|>Deep image-based Adaptive BRDF Measure: Efficient and accurate measurement of the bi-directional reflectance distribution function (BRDF) plays a key role in high quality image rendering and physically accurate sensor simulation. However, obtaining the reflectance properties of a material is both time-consuming and challenging. This paper presents a novel method for minimizing the number of samples required for high quality BRDF capture using a gonio-reflectometer setup. Taking an image of the physical material sample as input a lightweight neural network first estimates the parameters of an analytic BRDF model, and the distribution of the sample locations. In a second step we use an image based loss to find the number of samples required to meet the accuracy required. This approach significantly accelerates the measurement process while maintaining a high level of accuracy and fidelity in the BRDF representation.<|reference_end|> | arxiv | @article{cao2024deep,
title={Deep image-based Adaptive BRDF Measure},
author={Wen Cao},
journal={arXiv preprint arXiv:2410.02917},
year={2024},
archivePrefix={arXiv},
eprint={2410.02917},
primaryClass={cs.GR cs.AI}
} | cao2024deep |
arxiv-665321 | 2410.02921 | AirLetters: An Open Video Dataset of Characters Drawn in the Air | <|reference_start|>AirLetters: An Open Video Dataset of Characters Drawn in the Air: We introduce AirLetters, a new video dataset consisting of real-world videos of human-generated, articulated motions. Specifically, our dataset requires a vision model to predict letters that humans draw in the air. Unlike existing video datasets, accurate classification predictions for AirLetters rely critically on discerning motion patterns and on integrating long-range information in the video over time. An extensive evaluation of state-of-the-art image and video understanding models on AirLetters shows that these methods perform poorly and fall far behind a human baseline. Our work shows that, despite recent progress in end-to-end video understanding, accurate representations of complex articulated motions -- a task that is trivial for humans -- remains an open problem for end-to-end learning.<|reference_end|> | arxiv | @article{dagli2024airletters:,
title={AirLetters: An Open Video Dataset of Characters Drawn in the Air},
author={Rishit Dagli and Guillaume Berger and Joanna Materzynska and Ingo Bax
and Roland Memisevic},
journal={arXiv preprint arXiv:2410.02921},
year={2024},
archivePrefix={arXiv},
eprint={2410.02921},
primaryClass={cs.CV}
} | dagli2024airletters: |
arxiv-665322 | 2410.02923 | Random vortex and expansion-rate model for Oberbeck-Boussinesq fluid flows | <|reference_start|>Random vortex and expansion-rate model for Oberbeck-Boussinesq fluid flows: By using a formulation of a class of compressible viscous flows with a heat source via vorticity and expansion-rate, we study the Oberbeck-Boussinesq flows. To this end we establish a new integral representation for solutions of parabolic equations subject to certain boundary condition, which allows us to develop a random vortex method for certain compressible flows and to compute numerically solutions of their dynamical models. Numerical experiments are carried out, which not only capture detailed B\'enard convection but also are capable of providing additional information on the fluid density and the dynamics of expansion-rate of the flow.<|reference_end|> | arxiv | @article{guo2024random,
title={Random vortex and expansion-rate model for Oberbeck-Boussinesq fluid
flows},
author={Zihao Guo and Zhongmin Qian and Zihao Shen},
journal={arXiv preprint arXiv:2410.02923},
year={2024},
archivePrefix={arXiv},
eprint={2410.02923},
primaryClass={math.AP cs.NA math.NA math.PR physics.flu-dyn}
} | guo2024random |
arxiv-665323 | 2410.02924 | RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions | <|reference_start|>RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions: We propose a method for metric-scale monocular depth estimation. Inferring depth from a single image is an ill-posed problem due to the loss of scale from perspective projection during the image formation process. Any scale chosen is a bias, typically stemming from training on a dataset; hence, existing works have instead opted to use relative (normalized, inverse) depth. Our goal is to recover metric-scaled depth maps through a linear transformation. The crux of our method lies in the observation that certain objects (e.g., cars, trees, street signs) are typically found or associated with certain types of scenes (e.g., outdoor). We explore whether language descriptions can be used to transform relative depth predictions to those in metric scale. Our method, RSA, takes as input a text caption describing objects present in an image and outputs the parameters of a linear transformation which can be applied globally to a relative depth map to yield metric-scaled depth predictions. We demonstrate our method on recent general-purpose monocular depth models on indoors (NYUv2) and outdoors (KITTI). When trained on multiple datasets, RSA can serve as a general alignment module in zero-shot settings. Our method improves over common practices in aligning relative to metric depth and results in predictions that are comparable to an upper bound of fitting relative depth to ground truth via a linear transformation.<|reference_end|> | arxiv | @article{zeng2024rsa:,
title={RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through
Language Descriptions},
author={Ziyao Zeng, Yangchao Wu, Hyoungseob Park, Daniel Wang, Fengyu Yang,
Stefano Soatto, Dong Lao, Byung-Woo Hong, Alex Wong},
journal={arXiv preprint arXiv:2410.02924},
year={2024},
archivePrefix={arXiv},
eprint={2410.02924},
primaryClass={cs.CV}
} | zeng2024rsa: |
arxiv-665324 | 2410.02925 | A second order finite volume IMEX Runge-Kutta scheme for two dimensional PDEs in finance | <|reference_start|>A second order finite volume IMEX Runge-Kutta scheme for two dimensional PDEs in finance: In this article we present a novel and general methodology for building second order finite volume implicit-explicit (IMEX) numerical schemes for solving two dimensional financial parabolic PDEs with mixed derivatives. In particular, applications to basket and Heston models are presented. The obtained numerical schemes have excellent properties and are able to overcome the well-documented difficulties related with numerical approximations in the financial literature. The methods achieve true second order convergence with non-regular initial conditions. Besides, the IMEX time integrator allows to overcome the tiny time-step induced by the diffusive term in the explicit schemes, also providing very accurate and non-oscillatory approximations of the Greeks. Finally, in order to assess all the aforementioned good properties of the developed numerical schemes, we compute extremely accurate semi-analytic solutions using multi-dimensional Fourier cosine expansions. A novel technique to truncate the Fourier series for basket options is presented and it is efficiently implemented using multi-GPUs.<|reference_end|> | arxiv | @article{lópez-salas2024a,
title={A second order finite volume IMEX Runge-Kutta scheme for two dimensional
PDEs in finance},
author={J. G. L'opez-Salas, M. Su'arez-Taboada, M. J. Castro, A. M.
Ferreiro-Ferreiro, J. A. Garc'ia-Rodr'iguez},
journal={arXiv preprint arXiv:2410.02925},
year={2024},
archivePrefix={arXiv},
eprint={2410.02925},
primaryClass={math.NA cs.NA q-fin.CP q-fin.MF}
} | lópez-salas2024a |
arxiv-665325 | 2410.02927 | Boundary treatment for high-order IMEX Runge-Kutta local discontinuous Galerkin schemes for multidimensional nonlinear parabolic PDEs | <|reference_start|>Boundary treatment for high-order IMEX Runge-Kutta local discontinuous Galerkin schemes for multidimensional nonlinear parabolic PDEs: In this article, we propose novel boundary treatment algorithms to avoid order reduction when implicit-explicit Runge-Kutta time discretization is used for solving convection-diffusion-reaction problems with time-dependent Di\-richlet boundary conditions. We consider Cartesian meshes and PDEs with stiff terms coming from the diffusive parts of the PDE. The algorithms treat boundary values at the implicit-explicit internal stages in the same way as the interior points. The boundary treatment strategy is designed to work with multidimensional problems with possible nonlinear advection and source terms. The proposed methods recover the designed order of convergence by numerical verification. For the spatial discretization, in this work, we consider Local Discontinuous Galerkin methods, although the developed boundary treatment algorithms can operate with other discretization schemes in space, such as Finite Differences, Finite Elements or Finite Volumes.<|reference_end|> | arxiv | @article{gonzález-tabernero2024boundary,
title={Boundary treatment for high-order IMEX Runge-Kutta local discontinuous
Galerkin schemes for multidimensional nonlinear parabolic PDEs},
author={V. Gonz'alez-Tabernero, J. G. L'opez-Salas, M. J. Castro-D'iaz, J.
A. Garc'ia-Rodr'iguez},
journal={arXiv preprint arXiv:2410.02927},
year={2024},
archivePrefix={arXiv},
eprint={2410.02927},
primaryClass={math.NA cs.NA q-fin.CP q-fin.MF}
} | gonzález-tabernero2024boundary |
arxiv-665326 | 2410.02930 | Graph-tree Fusion Model with Bidirectional Information Propagation for Long Document Classification | <|reference_start|>Graph-tree Fusion Model with Bidirectional Information Propagation for Long Document Classification: Long document classification presents challenges in capturing both local and global dependencies due to their extensive content and complex structure. Existing methods often struggle with token limits and fail to adequately model hierarchical relationships within documents. To address these constraints, we propose a novel model leveraging a graph-tree structure. Our approach integrates syntax trees for sentence encodings and document graphs for document encodings, which capture fine-grained syntactic relationships and broader document contexts, respectively. We use Tree Transformers to generate sentence encodings, while a graph attention network models inter- and intra-sentence dependencies. During training, we implement bidirectional information propagation from word-to-sentence-to-document and vice versa, which enriches the contextual representation. Our proposed method enables a comprehensive understanding of content at all hierarchical levels and effectively handles arbitrarily long contexts without token limit constraints. Experimental results demonstrate the effectiveness of our approach in all types of long document classification tasks.<|reference_end|> | arxiv | @article{roy2024graph-tree,
title={Graph-tree Fusion Model with Bidirectional Information Propagation for
Long Document Classification},
author={Sudipta Singha Roy, Xindi Wang, Robert E. Mercer, Frank Rudzicz},
journal={arXiv preprint arXiv:2410.02930},
year={2024},
archivePrefix={arXiv},
eprint={2410.02930},
primaryClass={cs.CL}
} | roy2024graph-tree |
arxiv-665327 | 2410.02932 | Intrinsic Evaluation of RAG Systems for Deep-Logic Questions | <|reference_start|>Intrinsic Evaluation of RAG Systems for Deep-Logic Questions: We introduce the Overall Performance Index (OPI), an intrinsic metric to evaluate retrieval-augmented generation (RAG) mechanisms for applications involving deep-logic queries. OPI is computed as the harmonic mean of two key metrics: the Logical-Relation Correctness Ratio and the average of BERT embedding similarity scores between ground-truth and generated answers. We apply OPI to assess the performance of LangChain, a popular RAG tool, using a logical relations classifier fine-tuned from GPT-4o on the RAG-Dataset-12000 from Hugging Face. Our findings show a strong correlation between BERT embedding similarity scores and extrinsic evaluation scores. Among the commonly used retrievers, the cosine similarity retriever using BERT-based embeddings outperforms others, while the Euclidean distance-based retriever exhibits the weakest performance. Furthermore, we demonstrate that combining multiple retrievers, either algorithmically or by merging retrieved sentences, yields superior performance compared to using any single retriever alone.<|reference_end|> | arxiv | @article{hu2024intrinsic,
title={Intrinsic Evaluation of RAG Systems for Deep-Logic Questions},
author={Junyi Hu, You Zhou, Jie Wang},
journal={arXiv preprint arXiv:2410.02932},
year={2024},
archivePrefix={arXiv},
eprint={2410.02932},
primaryClass={cs.AI}
} | hu2024intrinsic |
arxiv-665328 | 2410.02935 | On Expert Estimation in Hierarchical Mixture of Experts: Beyond Softmax Gating Functions | <|reference_start|>On Expert Estimation in Hierarchical Mixture of Experts: Beyond Softmax Gating Functions: With the growing prominence of the Mixture of Experts (MoE) architecture in developing large-scale foundation models, we investigate the Hierarchical Mixture of Experts (HMoE), a specialized variant of MoE that excels in handling complex inputs and improving performance on targeted tasks. Our investigation highlights the advantages of using varied gating functions, moving beyond softmax gating within HMoE frameworks. We theoretically demonstrate that applying tailored gating functions to each expert group allows HMoE to achieve robust results, even when optimal gating functions are applied only at select hierarchical levels. Empirical validation across diverse scenarios supports these theoretical claims. This includes large-scale multimodal tasks, image classification, and latent domain discovery and prediction tasks, where our modified HMoE models show great performance improvements.<|reference_end|> | arxiv | @article{nguyen2024on,
title={On Expert Estimation in Hierarchical Mixture of Experts: Beyond Softmax
Gating Functions},
author={Huy Nguyen and Xing Han and Carl William Harris and Suchi Saria and
Nhat Ho},
journal={arXiv preprint arXiv:2410.02935},
year={2024},
archivePrefix={arXiv},
eprint={2410.02935},
primaryClass={stat.ML cs.LG}
} | nguyen2024on |
arxiv-665329 | 2410.02937 | Comparison of Autoencoder Encodings for ECG Representation in Downstream Prediction Tasks | <|reference_start|>Comparison of Autoencoder Encodings for ECG Representation in Downstream Prediction Tasks: The electrocardiogram (ECG) is an inexpensive and widely available tool for cardiovascular assessment. Despite its standardized format and small file size, the high complexity and inter-individual variability of ECG signals (typically a 60,000-size vector) make it challenging to use in deep learning models, especially when only small datasets are available. This study addresses these challenges by exploring feature generation methods from representative beat ECGs, focusing on Principal Component Analysis (PCA) and Autoencoders to reduce data complexity. We introduce three novel Variational Autoencoder (VAE) variants: Stochastic Autoencoder (SAE), Annealed beta-VAE (Abeta-VAE), and cyclical beta-VAE (Cbeta-VAE), and compare their effectiveness in maintaining signal fidelity and enhancing downstream prediction tasks. The Abeta-VAE achieved superior signal reconstruction, reducing the mean absolute error (MAE) to 15.7 plus-minus 3.2 microvolts, which is at the level of signal noise. Moreover, the SAE encodings, when combined with ECG summary features, improved the prediction of reduced Left Ventricular Ejection Fraction (LVEF), achieving an area under the receiver operating characteristic curve (AUROC) of 0.901. This performance nearly matches the 0.910 AUROC of state-of-the-art CNN models but requires significantly less data and computational resources. Our findings demonstrate that these VAE encodings are not only effective in simplifying ECG data but also provide a practical solution for applying deep learning in contexts with limited-scale labeled training data.<|reference_end|> | arxiv | @article{harvey2024comparison,
title={Comparison of Autoencoder Encodings for ECG Representation in Downstream
Prediction Tasks},
author={Christopher J. Harvey, Sumaiya Shomaji, Zijun Yao, Amit Noheria},
journal={arXiv preprint arXiv:2410.02937},
year={2024},
archivePrefix={arXiv},
eprint={2410.02937},
primaryClass={cs.LG eess.SP}
} | harvey2024comparison |
arxiv-665330 | 2410.02939 | Inductive Generative Recommendation via Retrieval-based Speculation | <|reference_start|>Inductive Generative Recommendation via Retrieval-based Speculation: Generative recommendation (GR) is an emerging paradigm that tokenizes items into discrete tokens and learns to autoregressively generate the next tokens as predictions. Although effective, GR models operate in a transductive setting, meaning they can only generate items seen during training without applying heuristic re-ranking strategies. In this paper, we propose SpecGR, a plug-and-play framework that enables GR models to recommend new items in an inductive setting. SpecGR uses a drafter model with inductive capability to propose candidate items, which may include both existing items and new items. The GR model then acts as a verifier, accepting or rejecting candidates while retaining its strong ranking capabilities. We further introduce the guided re-drafting technique to make the proposed candidates more aligned with the outputs of generative recommendation models, improving the verification efficiency. We consider two variants for drafting: (1) using an auxiliary drafter model for better flexibility, or (2) leveraging the GR model's own encoder for parameter-efficient self-drafting. Extensive experiments on three real-world datasets demonstrate that SpecGR exhibits both strong inductive recommendation ability and the best overall performance among the compared methods. Our code is available at: https://github.com/Jamesding000/SpecGR.<|reference_end|> | arxiv | @article{ding2024inductive,
title={Inductive Generative Recommendation via Retrieval-based Speculation},
author={Yijie Ding, Yupeng Hou, Jiacheng Li, Julian McAuley},
journal={arXiv preprint arXiv:2410.02939},
year={2024},
archivePrefix={arXiv},
eprint={2410.02939},
primaryClass={cs.IR}
} | ding2024inductive |
arxiv-665331 | 2410.02942 | SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric Groups | <|reference_start|>SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric Groups: Finite symmetric groups $S_n$ are essential in fields such as combinatorics, physics, and chemistry. However, learning a probability distribution over $S_n$ poses significant challenges due to its intractable size and discrete nature. In this paper, we introduce SymmetricDiffusers, a novel discrete diffusion model that simplifies the task of learning a complicated distribution over $S_n$ by decomposing it into learning simpler transitions of the reverse diffusion using deep neural networks. We identify the riffle shuffle as an effective forward transition and provide empirical guidelines for selecting the diffusion length based on the theory of random walks on finite groups. Additionally, we propose a generalized Plackett-Luce (PL) distribution for the reverse transition, which is provably more expressive than the PL distribution. We further introduce a theoretically grounded "denoising schedule" to improve sampling and learning efficiency. Extensive experiments show that our model achieves state-of-the-art or comparable performances on solving tasks including sorting 4-digit MNIST images, jigsaw puzzles, and traveling salesman problems. Our code is released at https://github.com/NickZhang53/SymmetricDiffusers.<|reference_end|> | arxiv | @article{zhang2024symmetricdiffusers:,
title={SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric
Groups},
author={Yongxing Zhang, Donglin Yang, Renjie Liao},
journal={arXiv preprint arXiv:2410.02942},
year={2024},
archivePrefix={arXiv},
eprint={2410.02942},
primaryClass={cs.LG cs.AI cs.CV}
} | zhang2024symmetricdiffusers: |
arxiv-665332 | 2410.02950 | LLMCO2: Advancing Accurate Carbon Footprint Prediction for LLM Inferences | <|reference_start|>LLMCO2: Advancing Accurate Carbon Footprint Prediction for LLM Inferences: Throughout its lifecycle, a large language model (LLM) generates a substantially larger carbon footprint during inference than training. LLM inference requests vary in batch size, prompt length, and token generation number, while cloud providers employ different GPU types and quantities to meet diverse service-level objectives for accuracy and latency. It is crucial for both users and cloud providers to have a tool that quickly and accurately estimates the carbon impact of LLM inferences based on a combination of inference request and hardware configurations before execution. Estimating the carbon footprint of LLM inferences is more complex than training due to lower and highly variable model FLOPS utilization, rendering previous equation-based models inaccurate. Additionally, existing machine learning (ML) prediction methods either lack accuracy or demand extensive training data, as they inadequately handle the distinct prefill and decode phases, overlook hardware-specific features, and inefficiently sample uncommon inference configurations. We introduce \coo, a graph neural network (GNN)-based model that greatly improves the accuracy of LLM inference carbon footprint predictions compared to previous methods.<|reference_end|> | arxiv | @article{fu2024llmco2:,
title={LLMCO2: Advancing Accurate Carbon Footprint Prediction for LLM
Inferences},
author={Zhenxiao Fu, Fan Chen, Shan Zhou, Haitong Li, Lei Jiang},
journal={arXiv preprint arXiv:2410.02950},
year={2024},
archivePrefix={arXiv},
eprint={2410.02950},
primaryClass={cs.LG cs.AI cs.CL cs.CY}
} | fu2024llmco2: |
arxiv-665333 | 2410.02952 | Visual Editing with LLM-based Tool Chaining: An Efficient Distillation Approach for Real-Time Applications | <|reference_start|>Visual Editing with LLM-based Tool Chaining: An Efficient Distillation Approach for Real-Time Applications: We present a practical distillation approach to fine-tune LLMs for invoking tools in real-time applications. We focus on visual editing tasks; specifically, we modify images and videos by interpreting user stylistic requests, specified in natural language ("golden hour"), using an LLM to select the appropriate tools and their parameters to achieve the desired visual effect. We found that proprietary LLMs such as GPT-3.5-Turbo show potential in this task, but their high cost and latency make them unsuitable for real-time applications. In our approach, we fine-tune a (smaller) student LLM with guidance from a (larger) teacher LLM and behavioral signals. We introduce offline metrics to evaluate student LLMs. Both online and offline experiments show that our student models manage to match the performance of our teacher model (GPT-3.5-Turbo), significantly reducing costs and latency. Lastly, we show that fine-tuning was improved by 25% in low-data regimes using augmentation.<|reference_end|> | arxiv | @article{sultan2024visual,
title={Visual Editing with LLM-based Tool Chaining: An Efficient Distillation
Approach for Real-Time Applications},
author={Oren Sultan, Alex Khasin, Guy Shiran, Asnat Greenstein-Messica, Dafna
Shahaf},
journal={arXiv preprint arXiv:2410.02952},
year={2024},
archivePrefix={arXiv},
eprint={2410.02952},
primaryClass={cs.CL cs.AI}
} | sultan2024visual |
arxiv-665334 | 2410.02953 | Unlocking Structured Thinking in Language Models with Cognitive prompting | <|reference_start|>Unlocking Structured Thinking in Language Models with Cognitive prompting: We propose cognitive prompting as a novel approach to guide problem-solving in large language models (LLMs) through structured, human-like cognitive operations such as goal clarification, decomposition, filtering, abstraction, and pattern recognition. By employing systematic, step-by-step reasoning, cognitive prompting enables LLMs to efficiently tackle complex, multi-step tasks. We evaluate the effectiveness of cognitive prompting on Meta's LLaMA models, comparing performance on arithmetic reasoning tasks using the GSM8K dataset and on commonsense reasoning benchmarks. Our analysis includes comparisons between models without cognitive prompting, models with a static sequence of cognitive operations, and models using reflective cognitive prompting, where the LLM dynamically self-selects the sequence of cognitive operations. The results show that cognitive prompting, particularly when dynamically adapted, significantly improves the performance of larger models, such as LLaMA3.1 70B, and enhances their ability to handle multi-step reasoning tasks. This approach also improves interpretability and flexibility, highlighting cognitive prompting as a promising strategy for general-purpose AI reasoning.<|reference_end|> | arxiv | @article{kramer2024unlocking,
title={Unlocking Structured Thinking in Language Models with Cognitive
Prompting},
author={Oliver Kramer, Jill Baumann},
journal={arXiv preprint arXiv:2410.02953},
year={2024},
archivePrefix={arXiv},
eprint={2410.02953},
primaryClass={cs.CL}
} | kramer2024unlocking |
arxiv-665335 | 2410.02954 | Digital Twin for O-RAN Towards 6G | <|reference_start|>Digital Twin for O-RAN Towards 6G: In future wireless systems of beyond 5G and 6G, addressing diverse applications with varying quality requirements is essential. Open Radio Access Network (O-RAN) architectures offer the potential for dynamic resource adaptation based on traffic demands. However, achieving real-time resource orchestration remains a challenge. Simultaneously, Digital Twin (DT) technology holds promise for testing and analysing complex systems, offering a unique platform for addressing dynamic operation and automation in O-RAN architectures. Yet, developing DTs for complex 5G/6G networks poses challenges, including data exchanges, ML model training data availability, network dynamics, processing power limitations, interdisciplinary collaboration needs, and a lack of standardized methodologies. This paper provides an overview of Open RAN architecture, trend and challenges, proposing the DT concepts for O-RAN with solution examples showcasing its integration into the framework.<|reference_end|> | arxiv | @article{nguyen2024digital,
title={Digital Twin for O-RAN Towards 6G},
author={Huan X. Nguyen, Kexuan Sun, Duc To, Quoc-Tuan Vien, Tuan Anh Le},
journal={arXiv preprint arXiv:2410.02954},
year={2024},
archivePrefix={arXiv},
eprint={2410.02954},
primaryClass={cs.NI cs.ET eess.SP}
} | nguyen2024digital |
arxiv-665336 | 2410.02955 | AiBAT: Artificial Intelligence/Instructions for Build, Assembly, and Test | <|reference_start|>AiBAT: Artificial Intelligence/Instructions for Build, Assembly, and Test: Instructions for Build, Assembly, and Test (IBAT) refers to the process used whenever any operation is conducted on hardware, including tests, assembly, and maintenance. Currently, the generation of IBAT documents is time-intensive, as users must manually reference and transfer information from engineering diagrams and parts lists into IBAT instructions. With advances in machine learning and computer vision, however, it is possible to have an artificial intelligence (AI) model perform the partial filling of the IBAT template, freeing up engineer time for more highly skilled tasks. AiBAT is a novel system for assisting users in authoring IBATs. It works by first analyzing assembly drawing documents, extracting information and parsing it, and then filling in IBAT templates with the extracted information. Such assisted authoring has potential to save time and reduce cost. This paper presents an overview of the AiBAT system, including promising preliminary results and discussion on future work.<|reference_end|> | arxiv | @article{nuernberger2024aibat:,
title={AiBAT: Artificial Intelligence/Instructions for Build, Assembly, and
Test},
author={Benjamin Nuernberger, Anny Liu, Heather Stefanini, Richard Otis,
Amanda Towler, R. Peter Dillon},
journal={arXiv preprint arXiv:2410.02955},
year={2024},
archivePrefix={arXiv},
eprint={2410.02955},
primaryClass={cs.AI cs.AR cs.ET cs.HC}
} | nuernberger2024aibat: |
arxiv-665337 | 2410.02956 | A System for Critical Facility and Resource Optimization in Disaster Management and Planning | <|reference_start|>A System for Critical Facility and Resource Optimization in Disaster Management and Planning: Disruptions to medical infrastructure during disasters pose significant risks to critically ill patients with advanced chronic kidney disease or end-stage renal disease. To enhance patient access to dialysis treatment under such conditions, it is crucial to assess the vulnerabilities of critical care facilities to hazardous events. This study proposes optimization models for patient reallocation and the strategic placement of temporary medical facilities to bolster the resilience of the critical care system, with a focus on equitable outcomes. Utilizing human mobility data from Texas, we evaluate patient access to critical care and dialysis centers under simulated hazard scenarios. The proposed bio-inspired optimization model, based on the Ant Colony optimization method, efficiently reallocates patients to mitigate disrupted access to dialysis facilities. The model outputs offer valuable insights into patient and hospital preparedness for disasters. Overall, the study presents a data-driven, analytics-based decision support tool designed to proactively mitigate potential disruptions in access to critical care facilities during disasters, tailored to the needs of health officials, emergency managers, and hospital system administrators in both the private and public sectors.<|reference_end|> | arxiv | @article{tung2024a,
title={A System for Critical Facility and Resource Optimization in Disaster
Management and Planning},
author={Emmanuel Tung, Ali Mostafavi, Maoxu Li, Sophie Li, Zeeshan Rasheed,
Khurram Shafique},
journal={arXiv preprint arXiv:2410.02956},
year={2024},
archivePrefix={arXiv},
eprint={2410.02956},
primaryClass={cs.NE}
} | tung2024a |
arxiv-665338 | 2410.02957 | Human Balancing on a Log: A Switched Multi-Layer Controller | <|reference_start|>Human Balancing on a Log: A Switched Multi-Layer Controller: We study the task of balancing a human on a log that is fixed in place. Balancing on a log is substantially more challenging than balancing on a flat surface -- to achieve stability, we use a switched multi-layer controller. The controller consists of an upper-layer LQR planner (akin to the central nervous system) that coordinates ankle and hip torques, and lower-layer PID trackers (akin to local motor units) that follow this plan subject to nonlinear dynamics. Additionally, the controller switches between three operational modes depending on the current state of the human. The efficacy of the controller is verified in simulation, where our controller is able to stabilize the human for a variety of initial conditions. We also show that this controller is compatible with muscle-based actuation and imperfect sensing, making it a promising candidate for modeling motor control under challenging conditions in a more bio-realistic way.<|reference_end|> | arxiv | @article{zhao2024human,
title={Human Balancing on a Log: A Switched Multi-Layer Controller},
author={Jiayi Zhao, Mo Yang, Jing Shuang Li},
journal={arXiv preprint arXiv:2410.02957},
year={2024},
archivePrefix={arXiv},
eprint={2410.02957},
primaryClass={eess.SY cs.SY}
} | zhao2024human |
arxiv-665339 | 2410.02958 | AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML | <|reference_start|>AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML: Automated machine learning (AutoML) accelerates AI development by automating tasks in the development pipeline, such as optimal model search and hyperparameter tuning. Existing AutoML systems often require technical expertise to set up complex tools, which is in general time-consuming and requires a large amount of human effort. Therefore, recent works have started exploiting large language models (LLM) to lessen such burden and increase the usability of AutoML frameworks via a natural language interface, allowing non-expert users to build their data-driven solutions. These methods, however, are usually designed only for a particular process in the AI development pipeline and do not efficiently use the inherent capacity of the LLMs. This paper proposes AutoML-Agent, a novel multi-agent framework tailored for full-pipeline AutoML, i.e., from data retrieval to model deployment. AutoML-Agent takes user's task descriptions, facilitates collaboration between specialized LLM agents, and delivers deployment-ready models. Unlike existing work, instead of devising a single plan, we introduce a retrieval-augmented planning strategy to enhance exploration to search for more optimal plans. We also decompose each plan into sub-tasks (e.g., data preprocessing and neural network design) each of which is solved by a specialized agent we build via prompting executing in parallel, making the search process more efficient. Moreover, we propose a multi-stage verification to verify executed results and guide the code generation LLM in implementing successful solutions. Extensive experiments on seven downstream tasks using fourteen datasets show that AutoML-Agent achieves a higher success rate in automating the full AutoML process, yielding systems with good performance throughout the diverse domains.<|reference_end|> | arxiv | @article{trirat2024automl-agent:,
title={AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML},
author={Patara Trirat, Wonyong Jeong, Sung Ju Hwang},
journal={arXiv preprint arXiv:2410.02958},
year={2024},
archivePrefix={arXiv},
eprint={2410.02958},
primaryClass={cs.LG cs.AI cs.CL cs.MA}
} | trirat2024automl-agent: |
arxiv-665340 | 2410.02959 | Coal Mining Question Answering with LLMs | <|reference_start|>Coal Mining Question Answering with LLMs: In this paper, we present a novel approach to coal mining question answering (QA) using large language models (LLMs) combined with tailored prompt engineering techniques. Coal mining is a complex, high-risk industry where accurate, context-aware information is critical for safe and efficient operations. Current QA systems struggle to handle the technical and dynamic nature of mining-related queries. To address these challenges, we propose a multi-turn prompt engineering framework designed to guide LLMs, such as GPT-4, in answering coal mining questions with higher precision and relevance. By breaking down complex queries into structured components, our approach allows LLMs to process nuanced technical information more effectively. We manually curated a dataset of 500 questions from real-world mining scenarios and evaluated the system's performance using both accuracy (ACC) and GPT-4-based scoring metrics. Experiments comparing ChatGPT, Claude2, and GPT-4 across baseline, chain-of-thought (CoT), and multi-turn prompting methods demonstrate that our method significantly improves both accuracy and contextual relevance, with an average accuracy improvement of 15-18\% and a notable increase in GPT-4 scores. The results show that our prompt-engineering approach provides a robust, adaptable solution for domain-specific question answering in high-stakes environments like coal mining.<|reference_end|> | arxiv | @article{rivera2024coal,
title={Coal Mining Question Answering with LLMs},
author={Antonio Carlos Rivera, Anthony Moore, Steven Robinson},
journal={arXiv preprint arXiv:2410.02959},
year={2024},
archivePrefix={arXiv},
eprint={2410.02959},
primaryClass={cs.CL}
} | rivera2024coal |
arxiv-665341 | 2410.02961 | LiDAR Inertial Odometry And Mapping Using Learned Registration-Relevant Features | <|reference_start|>LiDAR Inertial Odometry And Mapping Using Learned Registration-Relevant Features: SLAM is an important capability for many autonomous systems, and modern LiDAR-based methods offer promising performance. However, for long duration missions, existing works that either operate directly the full pointclouds or on extracted features face key tradeoffs in accuracy and computational efficiency (e.g., memory consumption). To address these issues, this paper presents DFLIOM with several key innovations. Unlike previous methods that rely on handcrafted heuristics and hand-tuned parameters for feature extraction, we propose a learning-based approach that select points relevant to LiDAR SLAM pointcloud registration. Furthermore, we extend our prior work DLIOM with the learned feature extractor and observe our method enables similar or even better localization performance using only about 20\% of the points in the dense point clouds. We demonstrate that DFLIOM performs well on multiple public benchmarks, achieving a 2.4\% decrease in localization error and 57.5\% decrease in memory usage compared to state-of-the-art methods (DLIOM). Although extracting features with the proposed network requires extra time, it is offset by the faster processing time downstream, thus maintaining real-time performance using 20Hz LiDAR on our hardware setup. The effectiveness of our learning-based feature extraction module is further demonstrated through comparison with several handcrafted feature extractors.<|reference_end|> | arxiv | @article{dong2024lidar,
title={LiDAR Inertial Odometry And Mapping Using Learned Registration-Relevant
Features},
author={Zihao Dong, Jeff Pflueger, Leonard Jung, David Thorne, Philip R.
Osteen, Christa S. Robison, Brett T. Lopez and Michael Everett},
journal={arXiv preprint arXiv:2410.02961},
year={2024},
archivePrefix={arXiv},
eprint={2410.02961},
primaryClass={cs.RO}
} | dong2024lidar |
arxiv-665342 | 2410.02963 | Bushfire Severity Modelling and Future Trend Prediction Across Australia: Integrating Remote Sensing and Machine Learning | <|reference_start|>Bushfire Severity Modelling and Future Trend Prediction Across Australia: Integrating Remote Sensing and Machine Learning: Bushfire is one of the major natural disasters that cause huge losses to livelihoods and the environment. Understanding and analyzing the severity of bushfires is crucial for effective management and mitigation strategies, helping to prevent the extensive damage and loss caused by these natural disasters. This study presents an in-depth analysis of bushfire severity in Australia over the last twelve years, combining remote sensing data and machine learning techniques to predict future fire trends. By utilizing Landsat imagery and integrating spectral indices like NDVI, NBR, and Burn Index, along with topographical and climatic factors, we developed a robust predictive model using XGBoost. The model achieved high accuracy, 86.13%, demonstrating its effectiveness in predicting fire severity across diverse Australian ecosystems. By analyzing historical trends and integrating factors such as population density and vegetation cover, we identify areas at high risk of future severe bushfires. Additionally, this research identifies key regions at risk, providing data-driven recommendations for targeted firefighting efforts. The findings contribute valuable insights into fire management strategies, enhancing resilience to future fire events in Australia. Also, we propose future work on developing a UAV-based swarm coordination model to enhance fire prediction in real-time and firefighting capabilities in the most vulnerable regions.<|reference_end|> | arxiv | @article{partheepan2024bushfire,
title={Bushfire Severity Modelling and Future Trend Prediction Across
Australia: Integrating Remote Sensing and Machine Learning},
author={Shouthiri Partheepan and Farzad Sanati and Jahan Hassan},
journal={arXiv preprint arXiv:2410.02963},
year={2024},
archivePrefix={arXiv},
eprint={2410.02963},
primaryClass={cs.CY cs.LG}
} | partheepan2024bushfire |
arxiv-665343 | 2410.02964 | A Simple Method for Secret-Key Generation Between Mobile Users Across Networks | <|reference_start|>A Simple Method for Secret-Key Generation Between Mobile Users Across Networks: Two or more mobiles users can continuously superimpose sequences of bits chosen from different packets or files already exchanged and authenticated between themselves to continuously renew a secret key for continuous strengthening of their privacy and authentication. This accumulative, adaptable and additive (AAA) method is discussed in this paper. The equivocation to Eve of any bit in the generated key by the AAA method equals to the probability that not all corresponding independent bits exchanged between the users are intercepted by Eve. This performance, achieved without using any knowledge of non-stationary probabilities of bits being intercepted by Eve, is compared to an established capacity achievable using that knowledge. A secrecy robustness of the AAA method against some correlations known to Eve is also discussed.<|reference_end|> | arxiv | @article{hua2024a,
title={A Simple Method for Secret-Key Generation Between Mobile Users Across
Networks},
author={Yingbo Hua},
journal={arXiv preprint arXiv:2410.02964},
year={2024},
archivePrefix={arXiv},
eprint={2410.02964},
primaryClass={cs.CR eess.SP}
} | hua2024a |
arxiv-665344 | 2410.02966 | Analyzing Fitts' Law using Offline and Online Optimal Control with Motor Noise | <|reference_start|>Analyzing Fitts' Law using Offline and Online Optimal Control with Motor Noise: The cause of the speed-accuracy tradeoff (typically quantified via Fitts' Law) is a debated topic of interest in motor neuroscience, and is commonly studied using tools from control theory. Two prominent theories involve the presence of signal dependent motor noise and planning variability -- these factors are generally incorporated separately. In this work, we study how well the simultaneous presence of both factors explains the speed-accuracy tradeoff. A human arm reaching model is developed with bio-realistic signal dependent motor noise, and a Gaussian noise model is used to deterministically approximate the motor noise. Both offline trajectory optimization and online model predictive control are used to simulate the planning and execution of several different reaching tasks with varying target sizes and movement durations. These reaching trajectories are then compared to experimental human reaching data, revealing that both models produce behavior consistent with humans, and the speed-accuracy tradeoff is present in both online and offline control. These results suggest the speed-accuracy tradeoff is likely caused by a combination of these two factors, and also that it plays a role in both offline and online computation.<|reference_end|> | arxiv | @article{bridges2024analyzing,
title={Analyzing Fitts' Law using Offline and Online Optimal Control with Motor
Noise},
author={Riley Bridges, Ethan Parham, Jing Shuang Li},
journal={arXiv preprint arXiv:2410.02966},
year={2024},
archivePrefix={arXiv},
eprint={2410.02966},
primaryClass={eess.SY cs.SY}
} | bridges2024analyzing |
arxiv-665345 | 2410.02967 | Label-Free Subjective Player Experience Modelling via Let's Play Videos | <|reference_start|>Label-Free Subjective Player Experience Modelling via Let's Play Videos: Player Experience Modelling (PEM) is the study of AI techniques applied to modelling a player's experience within a video game. PEM development can be labour-intensive, requiring expert hand-authoring or specialized data collection. In this work, we propose a novel PEM development approach, approximating player experience from gameplay video. We evaluate this approach predicting affect in the game Angry Birds via a human subject study. We validate that our PEM can strongly correlate with self-reported and sensor measures of affect, demonstrating the potential of this approach.<|reference_end|> | arxiv | @article{goel2024label-free,
title={Label-Free Subjective Player Experience Modelling via Let's Play Videos},
author={Dave Goel, Athar Mahmoudi-Nejad and Matthew Guzdial},
journal={arXiv preprint arXiv:2410.02967},
year={2024},
archivePrefix={arXiv},
eprint={2410.02967},
primaryClass={cs.HC cs.AI cs.LG}
} | goel2024label-free |
arxiv-665346 | 2410.02970 | F-Fidelity: A Robust Framework for Faithfulness Evaluation of Explainable AI | <|reference_start|>F-Fidelity: A Robust Framework for Faithfulness Evaluation of Explainable AI: Recent research has developed a number of eXplainable AI (XAI) techniques. Although extracting meaningful insights from deep learning models, how to properly evaluate these XAI methods remains an open problem. The most widely used approach is to perturb or even remove what the XAI method considers to be the most important features in an input and observe the changes in the output prediction. This approach although efficient suffers the Out-of-Distribution (OOD) problem as the perturbed samples may no longer follow the original data distribution. A recent method RemOve And Retrain (ROAR) solves the OOD issue by retraining the model with perturbed samples guided by explanations. However, the training may not always converge given the distribution difference. Furthermore, using the model retrained based on XAI methods to evaluate these explainers may cause information leakage and thus lead to unfair comparisons. We propose Fine-tuned Fidelity F-Fidelity, a robust evaluation framework for XAI, which utilizes i) an explanation-agnostic fine-tuning strategy, thus mitigating the information leakage issue and ii) a random masking operation that ensures that the removal step does not generate an OOD input. We designed controlled experiments with state-of-the-art (SOTA) explainers and their degraded version to verify the correctness of our framework. We conducted experiments on multiple data structures, such as images, time series, and natural language. The results demonstrate that F-Fidelity significantly improves upon prior evaluation metrics in recovering the ground-truth ranking of the explainers. Furthermore, we show both theoretically and empirically that, given a faithful explainer, F-Fidelity metric can be used to compute the sparsity of influential input components, i.e., to extract the true explanation size.<|reference_end|> | arxiv | @article{zheng2024f-fidelity:,
title={F-Fidelity: A Robust Framework for Faithfulness Evaluation of
Explainable AI},
author={Xu Zheng, Farhad Shirani, Zhuomin Chen, Chaohao Lin, Wei Cheng, Wenbo
Guo, Dongsheng Luo},
journal={arXiv preprint arXiv:2410.02970},
year={2024},
archivePrefix={arXiv},
eprint={2410.02970},
primaryClass={cs.LG cs.AI}
} | zheng2024f-fidelity: |
arxiv-665347 | 2410.02974 | Self-Deployable, Adaptive Soft Robots Based on Contracting-Cord Particle Jamming | <|reference_start|>Self-Deployable, Adaptive Soft Robots Based on Contracting-Cord Particle Jamming: We developed a new class of soft locomotive robots that can self-assemble into a preprogrammed configuration and vary their stiffness afterward in a highly integrated, compact body using contracting-cord particle jamming (CCPJ). We demonstrate this with a tripod-shaped robot, TripodBot, consisting of three CCPJ-based legs attached to a central body. TripodBot is intrinsically soft and can be stored and transported in a compact configuration. On site, it can self-deploy and crawl in a slip-stick manner through the shape morphing of its legs; a simplified analytical model accurately captures the speed. The robot's adaptability is demonstrated by its ability to navigate tunnels as narrow as 61 percent of its deployed body width and ceilings as low as 31 percent of its freestanding height. Additionally, it can climb slopes up to 15 degrees, carry a load of 5 grams (2.4 times its weight), and bear a load 9429 times its weight.<|reference_end|> | arxiv | @article{yan2024self-deployable,,
title={Self-Deployable, Adaptive Soft Robots Based on Contracting-Cord Particle
Jamming},
author={Wenzhong Yan, Brian Ye, Mingxi Li, Jonathan B. Hopkins, Ankur Mehta},
journal={arXiv preprint arXiv:2410.02974},
year={2024},
archivePrefix={arXiv},
eprint={2410.02974},
primaryClass={cs.RO}
} | yan2024self-deployable, |
arxiv-665348 | 2410.02976 | Learning Optimal Control and Dynamical Structure of Global Trajectory Search Problems with Diffusion Models | <|reference_start|>Learning Optimal Control and Dynamical Structure of Global Trajectory Search Problems with Diffusion Models: Spacecraft trajectory design is a global search problem, where previous work has revealed specific solution structures that can be captured with data-driven methods. This paper explores two global search problems in the circular restricted three-body problem: hybrid cost function of minimum fuel/time-of-flight and transfers to energy-dependent invariant manifolds. These problems display a fundamental structure either in the optimal control profile or the use of dynamical structures. We build on our prior generative machine learning framework to apply diffusion models to learn the conditional probability distribution of the search problem and analyze the model's capability to capture these structures.<|reference_end|> | arxiv | @article{graebner2024learning,
title={Learning Optimal Control and Dynamical Structure of Global Trajectory
Search Problems with Diffusion Models},
author={Jannik Graebner, Anjian Li, Amlan Sinha, Ryne Beeson},
journal={arXiv preprint arXiv:2410.02976},
year={2024},
archivePrefix={arXiv},
eprint={2410.02976},
primaryClass={cs.LG cs.SY eess.SY math.OC}
} | graebner2024learning |
arxiv-665349 | 2410.02977 | Harm Ratio: A Novel and Versatile Fairness Criterion | <|reference_start|>Harm Ratio: A Novel and Versatile Fairness Criterion: Envy-freeness has become the cornerstone of fair division research. In settings where each individual is allocated a disjoint share of collective resources, it is a compelling fairness axiom which demands that no individual strictly prefer the allocation of another individual to their own. Unfortunately, in many real-life collective decision-making problems, the goal is to choose a (common) public outcome that is equally applicable to all individuals, and the notion of envy becomes vacuous. Consequently, this literature has avoided studying fairness criteria that focus on individuals feeling a sense of jealousy or resentment towards other individuals (rather than towards the system), missing out on a key aspect of fairness. In this work, we propose a novel fairness criterion, individual harm ratio, which is inspired by envy-freeness but applies to a broad range of collective decision-making settings. Theoretically, we identify minimal conditions under which this criterion and its groupwise extensions can be guaranteed, and study the computational complexity of related problems. Empirically, we conduct experiments with real data to show that our fairness criterion is powerful enough to differentiate between prominent decision-making algorithms for a range of tasks from voting and fair division to participatory budgeting and peer review.<|reference_end|> | arxiv | @article{ebadian2024harm,
title={Harm Ratio: A Novel and Versatile Fairness Criterion},
author={Soroush Ebadian, Rupert Freeman, Nisarg Shah},
journal={arXiv preprint arXiv:2410.02977},
year={2024},
archivePrefix={arXiv},
eprint={2410.02977},
primaryClass={cs.GT cs.AI}
} | ebadian2024harm |
arxiv-665350 | 2410.02978 | An explainable approach to detect case law on housing and eviction issues within the HUDOC database | <|reference_start|>An explainable approach to detect case law on housing and eviction issues within the HUDOC database: Case law is instrumental in shaping our understanding of human rights, including the right to adequate housing. The HUDOC database provides access to the textual content of case law from the European Court of Human Rights (ECtHR), along with some metadata. While this metadata includes valuable information, such as the application number and the articles addressed in a case, it often lacks detailed substantive insights, such as the specific issues a case covers. This underscores the need for detailed analysis to extract such information. However, given the size of the database - containing over 40,000 cases - an automated solution is essential. In this study, we focus on the right to adequate housing and aim to build models to detect cases related to housing and eviction issues. Our experiments show that the resulting models not only provide performance comparable to more sophisticated approaches but are also interpretable, offering explanations for their decisions by highlighting the most influential words. The application of these models led to the identification of new cases that were initially overlooked during data collection. This suggests that NLP approaches can be effectively applied to categorise case law based on the specific issues they address.<|reference_end|> | arxiv | @article{mohammadi2024an,
title={An explainable approach to detect case law on housing and eviction
issues within the HUDOC database},
author={Mohammad Mohammadi, Martijn Wieling, Michel Vols},
journal={arXiv preprint arXiv:2410.02978},
year={2024},
archivePrefix={arXiv},
eprint={2410.02978},
primaryClass={cs.LG cs.AI}
} | mohammadi2024an |
arxiv-665351 | 2410.02979 | From Optimization to Sampling via Lyapunov Potentials | <|reference_start|>From Optimization to Sampling via Lyapunov Potentials: We study the problem of sampling from high-dimensional distributions using Langevin Dynamics, a natural and popular variant of Gradient Descent where at each step, appropriately scaled Gaussian noise is added. The similarities between Langevin Dynamics and Gradient Descent leads to the natural question: if the distribution's log-density can be optimized from all initializations via Gradient Descent, given oracle access to the gradients, can we sample from the distribution using Langevin Dynamics? We answer this question in the affirmative, at low but appropriate temperature levels natural in the context of both optimization and real-world applications. As a corollary, we show we can sample from several new natural and interesting classes of non-log-concave densities, an important setting where we have relatively few examples.<|reference_end|> | arxiv | @article{chen2024from,
title={From Optimization to Sampling via Lyapunov Potentials},
author={August Y. Chen, Karthik Sridharan},
journal={arXiv preprint arXiv:2410.02979},
year={2024},
archivePrefix={arXiv},
eprint={2410.02979},
primaryClass={stat.ML cs.LG math.ST stat.TH}
} | chen2024from |
arxiv-665352 | 2410.02980 | DecTrain: Deciding When to Train a DNN Online | <|reference_start|>DecTrain: Deciding When to Train a DNN Online: Deep neural networks (DNNs) can deteriorate in accuracy when deployment data differs from training data. While performing online training at all timesteps can improve accuracy, it is computationally expensive. We propose DecTrain, a new algorithm that decides when to train a monocular depth DNN online using self-supervision with low overhead. To make the decision at each timestep, DecTrain compares the cost of training with the predicted accuracy gain. We evaluate DecTrain on out-of-distribution data, and find DecTrain maintains accuracy compared to online training at all timesteps, while training only 44% of the time on average. We also compare the recovery of a low inference cost DNN using DecTrain and a more generalizable high inference cost DNN on various sequences. DecTrain recovers the majority (97%) of the accuracy gain of online training at all timesteps while reducing computation compared to the high inference cost DNN which recovers only 66%. With an even smaller DNN, we achieve 89% recovery while reducing computation by 56%. DecTrain enables low-cost online training for a smaller DNN to have competitive accuracy with a larger, more generalizable DNN at a lower overall computational cost.<|reference_end|> | arxiv | @article{fu2024dectrain:,
title={DecTrain: Deciding When to Train a DNN Online},
author={Zih-Sing Fu, Soumya Sudhakar, Sertac Karaman, Vivienne Sze},
journal={arXiv preprint arXiv:2410.02980},
year={2024},
archivePrefix={arXiv},
eprint={2410.02980},
primaryClass={cs.LG cs.RO}
} | fu2024dectrain: |
arxiv-665353 | 2410.02981 | GABIC: Graph-based Attention Block for Image Compression | <|reference_start|>GABIC: Graph-based Attention Block for Image Compression: While standardized codecs like JPEG and HEVC-intra represent the industry standard in image compression, neural Learned Image Compression (LIC) codecs represent a promising alternative. In detail, integrating attention mechanisms from Vision Transformers into LIC models has shown improved compression efficiency. However, extra efficiency often comes at the cost of aggregating redundant features. This work proposes a Graph-based Attention Block for Image Compression (GABIC), a method to reduce feature redundancy based on a k-Nearest Neighbors enhanced attention mechanism. Our experiments show that GABIC outperforms comparable methods, particularly at high bit rates, enhancing compression performance.<|reference_end|> | arxiv | @article{spadaro2024gabic:,
title={GABIC: Graph-based Attention Block for Image Compression},
author={Gabriele Spadaro, Alberto Presta, Enzo Tartaglione, Jhony H. Giraldo,
Marco Grangetto and Attilio Fiandrotti},
journal={arXiv preprint arXiv:2410.02981},
year={2024},
archivePrefix={arXiv},
eprint={2410.02981},
primaryClass={eess.IV cs.CV cs.LG}
} | spadaro2024gabic: |
arxiv-665354 | 2410.02983 | Information-Driven Search and Track of Novel Space Objects | <|reference_start|>Information-Driven Search and Track of Novel Space Objects: Space surveillance depends on efficiently directing sensor resources to maintain custody of known catalog objects. However, it remains unclear how to best utilize these resources to rapidly search for and track newly detected space objects. Provided a novel measurement, a search set can be instantiated through admissible region constraints to inform follow-up observations. In lacking well-constrained bounds, this set rapidly spreads in the along-track direction, growing much larger than a follow-up sensor's finite field of view. Moreover, the number of novel objects may be uncertain, and follow-up observations are most commonly corrupted by false positives from known catalog objects and missed detections. In this work, we address these challenges through the introduction of a joint sensor control and multi-target tracking approach. The search set associated to a novel measurement is represented by a Cardinalized Probability Hypothesis Density (CPHD), which jointly tracks the state uncertainty associated to a set of objects and a probability mass function for the true target number. In follow-up sensor scans, the information contained in an empty measurement set, and returns from both novel objects and known catalog objects is succinctly captured through this paradigm. To maximize the utility of a follow-up sensor, we introduce an information-driven sensor control approach for steering the instrument. Our methods are tested on two relevant test cases and we provide a comparative analysis with current naive tasking strategies.<|reference_end|> | arxiv | @article{wolf2024information-driven,
title={Information-Driven Search and Track of Novel Space Objects},
author={Trevor N. Wolf, Brandon A. Jones},
journal={arXiv preprint arXiv:2410.02983},
year={2024},
archivePrefix={arXiv},
eprint={2410.02983},
primaryClass={cs.RO cs.IT cs.SY eess.SY math.IT}
} | wolf2024information-driven |
arxiv-665355 | 2410.02984 | Differentiation and Specialization of Attention Heads via the Refined Local Learning Coefficient | <|reference_start|>Differentiation and Specialization of Attention Heads via the Refined Local Learning Coefficient: We introduce refined variants of the Local Learning Coefficient (LLC), a measure of model complexity grounded in singular learning theory, to study the development of internal structure in transformer language models during training. By applying these \textit{refined LLCs} (rLLCs) to individual components of a two-layer attention-only transformer, we gain novel insights into the progressive differentiation and specialization of attention heads. Our methodology reveals how attention heads differentiate into distinct functional roles over the course of training, analyzes the types of data these heads specialize to process, and discovers a previously unidentified multigram circuit. These findings demonstrate that rLLCs provide a principled, quantitative toolkit for \textit{developmental interpretability}, which aims to understand models through their evolution across the learning process. More broadly, this work takes a step towards establishing the correspondence between data distributional structure, geometric properties of the loss landscape, learning dynamics, and emergent computational structures in neural networks.<|reference_end|> | arxiv | @article{wang2024differentiation,
title={Differentiation and Specialization of Attention Heads via the Refined
Local Learning Coefficient},
author={George Wang, Jesse Hoogland, Stan van Wingerden, Zach Furman, Daniel
Murfet},
journal={arXiv preprint arXiv:2410.02984},
year={2024},
archivePrefix={arXiv},
eprint={2410.02984},
primaryClass={cs.LG cs.AI}
} | wang2024differentiation |
arxiv-665356 | 2410.02987 | Parrondo's effects with aperiodic protocols | <|reference_start|>Parrondo's effects with aperiodic protocols: In this work, we study the effectiveness of employing archetypal aperiodic sequencing -- namely Fibonacci, Thue-Morse, and Rudin-Saphiro -- on the Parrondian effect. From a capital gain perspective, our results show that these series do yield a Parrondo's Paradox with the Thue-Morse based strategy outperforming not only the other two aperiodic strategies but benchmark Parrondian games with random and periodical ($AABBAABB\ldots$) switching as well. The least performing of the three aperiodic strategies is the Rudin-Shapiro. To elucidate the underlying causes of these results, we analyze the cross-correlation between the capital generated by the switching protocols and that of the isolated losing games. This analysis reveals that a pronounced anti-correlation (below -0.95) with both isolated games is typically required to achieve a robust manifestation of Parrondo's effect. We also study the influence of the sequencing on the capital using the lacunarity and persistence measures. In general, we observe that the switching protocols tend to become less performing in terms of the capital as one increases the persistence and thus approaches the features of an isolated losing game. For the (log-)lacunarity, a property related to heterogeneity, we notice that for small persistence (less than 0.5) the performance increases with the lacunarity with a maximum around 0.4. In respect of this, our work shows that the optimisation of a switching protocol is strongly dependent on a fine tune between persistence and heterogeneity.<|reference_end|> | arxiv | @article{pires2024parrondo's,
title={Parrondo's effects with aperiodic protocols},
author={Marcelo A. Pires, Erveton P. Pinto, Rone N. da Silva, S'ilvio M.
Duarte Queir'os},
journal={arXiv preprint arXiv:2410.02987},
year={2024},
archivePrefix={arXiv},
eprint={2410.02987},
primaryClass={physics.soc-ph cs.GT stat.AP}
} | pires2024parrondo's |
arxiv-665357 | 2410.02988 | Fully Automated CTC Detection, Segmentation and Classification for Multi-Channel IF Imaging | <|reference_start|>Fully Automated CTC Detection, Segmentation and Classification for Multi-Channel IF Imaging: Liquid biopsies (eg., blood draws) offer a less invasive and non-localized alternative to tissue biopsies for monitoring the progression of metastatic breast cancer (mBCa). Immunofluoresence (IF) microscopy is a tool to image and analyze millions of blood cells in a patient sample. By detecting and genetically sequencing circulating tumor cells (CTCs) in the blood, personalized treatment plans are achievable for various cancer subtypes. However, CTCs are rare (about 1 in 2M), making manual CTC detection very difficult. In addition, clinicians rely on quantitative cellular biomarkers to manually classify CTCs. This requires prior tasks of cell detection, segmentation and feature extraction. To assist clinicians, we have developed a fully automated machine learning-based production-level pipeline to efficiently detect, segment and classify CTCs in multi-channel IF images. We achieve over 99% sensitivity and 97% specificity on 9,533 cells from 15 mBCa patients. Our pipeline has been successfully deployed on real mBCa patients, reducing a patient average of 14M detected cells to only 335 CTC candidates for manual review.<|reference_end|> | arxiv | @article{schwab2024fully,
title={Fully Automated CTC Detection, Segmentation and Classification for
Multi-Channel IF Imaging},
author={Evan Schwab, Bharat Annaldas, Nisha Ramesh, Anna Lundberg, Vishal
Shelke, Xinran Xu, Cole Gilbertson, Jiyun Byun, Ernest T. Lam},
journal={arXiv preprint arXiv:2410.02988},
year={2024},
archivePrefix={arXiv},
eprint={2410.02988},
primaryClass={cs.CV q-bio.QM}
} | schwab2024fully |
arxiv-665358 | 2410.02992 | Guided Stream of Search: Learning to Better Search with Language Models via Optimal Path Guidance | <|reference_start|>Guided Stream of Search: Learning to Better Search with Language Models via Optimal Path Guidance: While language models have demonstrated impressive capabilities across a range of tasks, they still struggle with tasks that require complex planning and reasoning. Recent studies have proposed training language models on search processes rather than optimal solutions, resulting in better generalization performance even though search processes are noisy and even suboptimal. However, these studies overlook the value of optimal solutions, which can serve as step-by-step landmarks to guide more effective search. In this work, we explore how to leverage optimal solutions to enhance the search and planning abilities of language models. To this end, we propose guided stream of search (GSoS), which seamlessly incorporates optimal solutions into the self-generation process in a progressive manner, producing high-quality search trajectories. These trajectories are then distilled into the pre-trained model via supervised fine-tuning. Our approach significantly enhances the search and planning abilities of language models on Countdown, a simple yet challenging mathematical reasoning task. Notably, combining our method with RL fine-tuning yields further improvements, whereas previous supervised fine-tuning methods do not benefit from RL. Furthermore, our approach exhibits greater effectiveness than leveraging optimal solutions in the form of subgoal rewards.<|reference_end|> | arxiv | @article{moon2024guided,
title={Guided Stream of Search: Learning to Better Search with Language Models
via Optimal Path Guidance},
author={Seungyong Moon, Bumsoo Park, Hyun Oh Song},
journal={arXiv preprint arXiv:2410.02992},
year={2024},
archivePrefix={arXiv},
eprint={2410.02992},
primaryClass={cs.AI cs.CL}
} | moon2024guided |
arxiv-665359 | 2410.02994 | Finite-Sample Analysis of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning | <|reference_start|>Finite-Sample Analysis of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning: Monte Carlo Exploring Starts (MCES), which aims to learn the optimal policy using only sample returns, is a simple and natural algorithm in reinforcement learning which has been shown to converge under various conditions. However, the convergence rate analysis for MCES-style algorithms in the form of sample complexity has received very little attention. In this paper we develop a finite sample bound for a modified MCES algorithm which solves the stochastic shortest path problem. To this end, we prove a novel result on the convergence rate of the policy iteration algorithm. This result implies that with probability at least $1-\delta$, the algorithm returns an optimal policy after $\tilde{O}(SAK^3\log^3\frac{1}{\delta})$ sampled episodes, where $S$ and $A$ denote the number of states and actions respectively, $K$ is a proxy for episode length, and $\tilde{O}$ hides logarithmic factors and constants depending on the rewards of the environment that are assumed to be known.<|reference_end|> | arxiv | @article{chen2024finite-sample,
title={Finite-Sample Analysis of the Monte Carlo Exploring Starts Algorithm for
Reinforcement Learning},
author={Suei-Wen Chen, Keith Ross and Pierre Youssef},
journal={arXiv preprint arXiv:2410.02994},
year={2024},
archivePrefix={arXiv},
eprint={2410.02994},
primaryClass={cs.LG}
} | chen2024finite-sample |
arxiv-665360 | 2410.02995 | Task-unaware Lifelong Robot Learning with Retrieval-based Weighted Local Adaptation | <|reference_start|>Task-unaware Lifelong Robot Learning with Retrieval-based Weighted Local Adaptation: Real-world environments require robots to continuously acquire new skills while retaining previously learned abilities, all without the need for clearly defined task boundaries. Storing all past data to prevent forgetting is impractical due to storage and privacy concerns. To address this, we propose a method that efficiently restores a robot's proficiency in previously learned tasks over its lifespan. Using an Episodic Memory (EM), our approach enables experience replay during training and retrieval during testing for local fine-tuning, allowing rapid adaptation to previously encountered problems without explicit task identifiers. Additionally, we introduce a selective weighting mechanism that emphasizes the most challenging segments of retrieved demonstrations, focusing local adaptation where it is most needed. This framework offers a scalable solution for lifelong learning in dynamic, task-unaware environments, combining retrieval-based adaptation with selective weighting to enhance robot performance in open-ended scenarios.<|reference_end|> | arxiv | @article{yang2024task-unaware,
title={Task-unaware Lifelong Robot Learning with Retrieval-based Weighted Local
Adaptation},
author={Pengzhi Yang, Xinyu Wang, Ruipeng Zhang, Cong Wang, Frans Oliehoek,
Jens Kober},
journal={arXiv preprint arXiv:2410.02995},
year={2024},
archivePrefix={arXiv},
eprint={2410.02995},
primaryClass={cs.RO cs.AI}
} | yang2024task-unaware |
arxiv-665361 | 2410.02998 | Q-SCALE: Quantum computing-based Sensor Calibration for Advanced Learning and Efficiency | <|reference_start|>Q-SCALE: Quantum computing-based Sensor Calibration for Advanced Learning and Efficiency: In a world burdened by air pollution, the integration of state-of-the-art sensor calibration techniques utilizing Quantum Computing (QC) and Machine Learning (ML) holds promise for enhancing the accuracy and efficiency of air quality monitoring systems in smart cities. This article investigates the process of calibrating inexpensive optical fine-dust sensors through advanced methodologies such as Deep Learning (DL) and Quantum Machine Learning (QML). The objective of the project is to compare four sophisticated algorithms from both the classical and quantum realms to discern their disparities and explore possible alternative approaches to improve the precision and dependability of particulate matter measurements in urban air quality surveillance. Classical Feed-Forward Neural Networks (FFNN) and Long Short-Term Memory (LSTM) models are evaluated against their quantum counterparts: Variational Quantum Regressors (VQR) and Quantum LSTM (QLSTM) circuits. Through meticulous testing, including hyperparameter optimization and cross-validation, the study assesses the potential of quantum models to refine calibration performance. Our analysis shows that: the FFNN model achieved superior calibration accuracy on the test set compared to the VQR model in terms of lower L1 loss function (2.92 vs 4.81); the QLSTM slightly outperformed the LSTM model (loss on the test set: 2.70 vs 2.77), despite using fewer trainable weights (66 vs 482).<|reference_end|> | arxiv | @article{bergadano2024q-scale:,
title={Q-SCALE: Quantum computing-based Sensor Calibration for Advanced
Learning and Efficiency},
author={Lorenzo Bergadano, Andrea Ceschini, Pietro Chiavassa, Edoardo Giusto,
Bartolomeo Montrucchio, Massimo Panella, Antonello Rosato},
journal={arXiv preprint arXiv:2410.02998},
year={2024},
archivePrefix={arXiv},
eprint={2410.02998},
primaryClass={cs.LG cs.ET}
} | bergadano2024q-scale: |
arxiv-665362 | 2410.03000 | Towards Universal Certified Robustness with Multi-Norm Training | <|reference_start|>Towards Universal Certified Robustness with Multi-Norm Training: Existing certified training methods can only train models to be robust against a certain perturbation type (e.g. $l_\infty$ or $l_2$). However, an $l_\infty$ certifiably robust model may not be certifiably robust against $l_2$ perturbation (and vice versa) and also has low robustness against other perturbations (e.g. geometric transformation). To this end, we propose the first multi-norm certified training framework \textbf{CURE}, consisting of a new $l_2$ deterministic certified training defense and several multi-norm certified training methods, to attain better \emph{union robustness} when training from scratch or fine-tuning a pre-trained certified model. Further, we devise bound alignment and connect natural training with certified training for better union robustness. Compared with SOTA certified training, \textbf{CURE} improves union robustness up to $22.8\%$ on MNIST, $23.9\%$ on CIFAR-10, and $8.0\%$ on TinyImagenet. Further, it leads to better generalization on a diverse set of challenging unseen geometric perturbations, up to $6.8\%$ on CIFAR-10. Overall, our contributions pave a path towards \textit{universal certified robustness}.<|reference_end|> | arxiv | @article{jiang2024towards,
title={Towards Universal Certified Robustness with Multi-Norm Training},
author={Enyi Jiang, Gagandeep Singh},
journal={arXiv preprint arXiv:2410.03000},
year={2024},
archivePrefix={arXiv},
eprint={2410.03000},
primaryClass={cs.LG cs.CR}
} | jiang2024towards |
arxiv-665363 | 2410.03001 | Can Transformers Learn $n$-gram Language Models? | <|reference_start|>Can Transformers Learn $n$-gram Language Models?: Much theoretical work has described the ability of transformers to represent formal languages. However, linking theoretical results to empirical performance is not straightforward due to the complex interplay between the architecture, the learning algorithm, and training data. To test whether theoretical lower bounds imply \emph{learnability} of formal languages, we turn to recent work relating transformers to $n$-gram language models (LMs). We study transformers' ability to learn random $n$-gram LMs of two kinds: ones with arbitrary next-symbol probabilities and ones where those are defined with shared parameters. We find that classic estimation techniques for $n$-gram LMs such as add-$\lambda$ smoothing outperform transformers on the former, while transformers perform better on the latter, outperforming methods specifically designed to learn $n$-gram LMs.<|reference_end|> | arxiv | @article{svete2024can,
title={Can Transformers Learn $n$-gram Language Models?},
author={Anej Svete, Nadav Borenstein, Mike Zhou, Isabelle Augenstein, Ryan
Cotterell},
journal={arXiv preprint arXiv:2410.03001},
year={2024},
archivePrefix={arXiv},
eprint={2410.03001},
primaryClass={cs.CL}
} | svete2024can |
arxiv-665364 | 2410.03006 | Formation of Representations in Neural Networks | <|reference_start|>Formation of Representations in Neural Networks: Understanding neural representations will help open the black box of neural networks and advance our scientific understanding of modern AI systems. However, how complex, structured, and transferable representations emerge in modern neural networks has remained a mystery. Building on previous results, we propose the Canonical Representation Hypothesis (CRH), which posits a set of six alignment relations to universally govern the formation of representations in most hidden layers of a neural network. Under the CRH, the latent representations (R), weights (W), and neuron gradients (G) become mutually aligned during training. This alignment implies that neural networks naturally learn compact representations, where neurons and weights are invariant to task-irrelevant transformations. We then show that the breaking of CRH leads to the emergence of reciprocal power-law relations between R, W, and G, which we refer to as the Polynomial Alignment Hypothesis (PAH). We present a minimal-assumption theory demonstrating that the balance between gradient noise and regularization is crucial for the emergence the canonical representation. The CRH and PAH lead to an exciting possibility of unifying major key deep learning phenomena, including neural collapse and the neural feature ansatz, in a single framework.<|reference_end|> | arxiv | @article{ziyin2024formation,
title={Formation of Representations in Neural Networks},
author={Liu Ziyin, Isaac Chuang, Tomer Galanti, Tomaso Poggio},
journal={arXiv preprint arXiv:2410.03006},
year={2024},
archivePrefix={arXiv},
eprint={2410.03006},
primaryClass={cs.LG cond-mat.dis-nn}
} | ziyin2024formation |
arxiv-665365 | 2410.03007 | FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model | <|reference_start|>FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model: In this study, we aim to explore Multitask Speech Language Model (SpeechLM) efficient inference via token reduction. Unlike other modalities such as vision or text, speech has unique temporal dependencies, making previous efficient inference works on other modalities not directly applicable. Furthermore, methods for efficient SpeechLM inference on long sequence and sparse signals remain largely unexplored. Then we propose FastAdaSP, a weighted token merging framework specifically designed for various speech-related tasks to improve the trade-off between efficiency and performance. Experimental results on WavLLM and Qwen-Audio show that our method achieves the state-of-the-art (SOTA) efficiency-performance trade-off compared with other baseline methods. Specifically, FastAdaSP achieved 7x memory efficiency and 1.83x decoding throughput without any degradation on tasks like Emotion Recognition (ER) and Spoken Question Answering (SQA). The code will be available at https://github.com/yichen14/FastAdaSP<|reference_end|> | arxiv | @article{lu2024fastadasp:,
title={FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech
Language Model},
author={Yichen Lu, Jiaqi Song, Chao-Han Huck Yang, Shinji Watanabe},
journal={arXiv preprint arXiv:2410.03007},
year={2024},
archivePrefix={arXiv},
eprint={2410.03007},
primaryClass={eess.AS cs.AI cs.CL}
} | lu2024fastadasp: |
arxiv-665366 | 2410.03010 | MMP: Towards Robust Multi-Modal Learning with Masked Modality Projection | <|reference_start|>MMP: Towards Robust Multi-Modal Learning with Masked Modality Projection: Multimodal learning seeks to combine data from multiple input sources to enhance the performance of different downstream tasks. In real-world scenarios, performance can degrade substantially if some input modalities are missing. Existing methods that can handle missing modalities involve custom training or adaptation steps for each input modality combination. These approaches are either tied to specific modalities or become computationally expensive as the number of input modalities increases. In this paper, we propose Masked Modality Projection (MMP), a method designed to train a single model that is robust to any missing modality scenario. We achieve this by randomly masking a subset of modalities during training and learning to project available input modalities to estimate the tokens for the masked modalities. This approach enables the model to effectively learn to leverage the information from the available modalities to compensate for the missing ones, enhancing missing modality robustness. We conduct a series of experiments with various baseline models and datasets to assess the effectiveness of this strategy. Experiments demonstrate that our approach improves robustness to different missing modality scenarios, outperforming existing methods designed for missing modalities or specific modality combinations.<|reference_end|> | arxiv | @article{nezakati2024mmp:,
title={MMP: Towards Robust Multi-Modal Learning with Masked Modality Projection},
author={Niki Nezakati, Md Kaykobad Reza, Ameya Patil, Mashhour Solh, M. Salman
Asif},
journal={arXiv preprint arXiv:2410.03010},
year={2024},
archivePrefix={arXiv},
eprint={2410.03010},
primaryClass={cs.LG cs.CV}
} | nezakati2024mmp: |
arxiv-665367 | 2410.03011 | Towards Understanding the Universality of Transformers for Next-Token Prediction | <|reference_start|>Towards Understanding the Universality of Transformers for Next-Token Prediction: Causal Transformers are trained to predict the next token for a given context. While it is widely accepted that self-attention is crucial for encoding the causal structure of sequences, the precise underlying mechanism behind this in-context autoregressive learning ability remains unclear. In this paper, we take a step towards understanding this phenomenon by studying the approximation ability of Transformers for next-token prediction. Specifically, we explore the capacity of causal Transformers to predict the next token $x_{t+1}$ given an autoregressive sequence $(x_1, \dots, x_t)$ as a prompt, where $ x_{t+1} = f(x_t) $, and $ f $ is a context-dependent function that varies with each sequence. On the theoretical side, we focus on specific instances, namely when $ f $ is linear or when $ (x_t)_{t \geq 1} $ is periodic. We explicitly construct a Transformer (with linear, exponential, or softmax attention) that learns the mapping $f$ in-context through a causal kernel descent method. The causal kernel descent method we propose provably estimates $x_{t+1} $ based solely on past and current observations $ (x_1, \dots, x_t) $, with connections to the Kaczmarz algorithm in Hilbert spaces. We present experimental results that validate our theoretical findings and suggest their applicability to more general mappings $f$.<|reference_end|> | arxiv | @article{sander2024towards,
title={Towards Understanding the Universality of Transformers for Next-Token
Prediction},
author={Michael E. Sander and Gabriel Peyr'e},
journal={arXiv preprint arXiv:2410.03011},
year={2024},
archivePrefix={arXiv},
eprint={2410.03011},
primaryClass={stat.ML cs.LG}
} | sander2024towards |
arxiv-665368 | 2410.03016 | Learning a Fast Mixing Exogenous Block MDP using a Single Trajectory | <|reference_start|>Learning a Fast Mixing Exogenous Block MDP using a Single Trajectory: In order to train agents that can quickly adapt to new objectives or reward functions, efficient unsupervised representation learning in sequential decision-making environments can be important. Frameworks such as the Exogenous Block Markov Decision Process (Ex-BMDP) have been proposed to formalize this representation-learning problem (Efroni et al., 2022b). In the Ex-BMDP framework, the agent's high-dimensional observations of the environment have two latent factors: a controllable factor, which evolves deterministically within a small state space according to the agent's actions, and an exogenous factor, which represents time-correlated noise, and can be highly complex. The goal of the representation learning problem is to learn an encoder that maps from observations into the controllable latent space, as well as the dynamics of this space. Efroni et al. (2022b) has shown that this is possible with a sample complexity that depends only on the size of the controllable latent space, and not on the size of the noise factor. However, this prior work has focused on the episodic setting, where the controllable latent state resets to a specific start state after a finite horizon. By contrast, if the agent can only interact with the environment in a single continuous trajectory, prior works have not established sample-complexity bounds. We propose STEEL, the first provably sample-efficient algorithm for learning the controllable dynamics of an Ex-BMDP from a single trajectory, in the function approximation setting. STEEL has a sample complexity that depends only on the sizes of the controllable latent space and the encoder function class, and (at worst linearly) on the mixing time of the exogenous noise factor. We prove that STEEL is correct and sample-efficient, and demonstrate STEEL on two toy problems. Code is available at: https://github.com/midi-lab/steel.<|reference_end|> | arxiv | @article{levine2024learning,
title={Learning a Fast Mixing Exogenous Block MDP using a Single Trajectory},
author={Alexander Levine, Peter Stone, Amy Zhang},
journal={arXiv preprint arXiv:2410.03016},
year={2024},
archivePrefix={arXiv},
eprint={2410.03016},
primaryClass={cs.LG}
} | levine2024learning |
arxiv-665369 | 2410.03017 | Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise | <|reference_start|>Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise: Generative AI, particularly Language Models (LMs), has the potential to transform real-world domains with societal impact, particularly where access to experts is limited. For example, in education, training novice educators with expert guidance is important for effectiveness but expensive, creating significant barriers to improving education quality at scale. This challenge disproportionately harms students from under-served communities, who stand to gain the most from high-quality education. We introduce Tutor CoPilot, a novel Human-AI approach that leverages a model of expert thinking to provide expert-like guidance to tutors as they tutor. This study is the first randomized controlled trial of a Human-AI system in live tutoring, involving 900 tutors and 1,800 K-12 students from historically under-served communities. Following a preregistered analysis plan, we find that students working with tutors that have access to Tutor CoPilot are 4 percentage points (p.p.) more likely to master topics (p<0.01). Notably, students of lower-rated tutors experienced the greatest benefit, improving mastery by 9 p.p. We find that Tutor CoPilot costs only $20 per-tutor annually. We analyze 550,000+ messages using classifiers to identify pedagogical strategies, and find that tutors with access to Tutor CoPilot are more likely to use high-quality strategies to foster student understanding (e.g., asking guiding questions) and less likely to give away the answer to the student. Tutor interviews highlight how Tutor CoPilot's guidance helps tutors to respond to student needs, though they flag issues in Tutor CoPilot, such as generating suggestions that are not grade-level appropriate. Altogether, our study of Tutor CoPilot demonstrates how Human-AI systems can scale expertise in real-world domains, bridge gaps in skills and create a future where high-quality education is accessible to all students.<|reference_end|> | arxiv | @article{wang2024tutor,
title={Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise},
author={Rose E. Wang, Ana T. Ribeiro, Carly D. Robinson, Susanna Loeb, Dora
Demszky},
journal={arXiv preprint arXiv:2410.03017},
year={2024},
archivePrefix={arXiv},
eprint={2410.03017},
primaryClass={cs.CL}
} | wang2024tutor |
arxiv-665370 | 2410.03018 | Transforming Teachers' Roles and Agencies in the Era of Generative AI: Perceptions, Acceptance, Knowledge, and Practices | <|reference_start|>Transforming Teachers' Roles and Agencies in the Era of Generative AI: Perceptions, Acceptance, Knowledge, and Practices: This paper explores the transformative impact of Generative Artificial Intelligence (GenAI) on teachers' roles and agencies in education, presenting a comprehensive framework that addresses teachers' perceptions, knowledge, acceptance, and practices of GenAI. As GenAI technologies, such as ChatGPT, become increasingly integrated into educational settings, teachers are required to adapt to evolving classroom dynamics, where AI plays a significant role in content creation, personalized learning, and student engagement. However, existing literature often treats these factors in isolation, overlooking how they collectively influence teachers' ability to effectively integrate GenAI into their pedagogical practices. This paper fills this gap by proposing a framework that categorizes teachers into four roles -- Observer, Adopter, Collaborator, and Innovator -- each representing different levels of GenAI engagement, outlining teachers' agencies in GenAI classrooms. By highlighting the need for continuous professional development and institutional support, we demonstrate how teachers can evolve from basic GenAI users to co-creators of knowledge alongside GenAI systems. The findings emphasize that for GenAI to reach its full educational potential, teachers must not only accept and understand its capabilities but also integrate it deeply into their teaching strategies. This study contributes to the growing literature on GenAI in education, offering practical implications for supporting teachers in navigating the complexities of GenAI adoption.<|reference_end|> | arxiv | @article{zhai2024transforming,
title={Transforming Teachers' Roles and Agencies in the Era of Generative AI:
Perceptions, Acceptance, Knowledge, and Practices},
author={Xiaoming Zhai},
journal={arXiv preprint arXiv:2410.03018},
year={2024},
archivePrefix={arXiv},
eprint={2410.03018},
primaryClass={cs.CY cs.AI}
} | zhai2024transforming |
arxiv-665371 | 2410.03019 | Is Your Paper Being Reviewed by an LLM? Investigating AI Text Detectability in Peer Review | <|reference_start|>Is Your Paper Being Reviewed by an LLM? Investigating AI Text Detectability in Peer Review: Peer review is a critical process for ensuring the integrity of published scientific research. Confidence in this process is predicated on the assumption that experts in the relevant domain give careful consideration to the merits of manuscripts which are submitted for publication. With the recent rapid advancements in the linguistic capabilities of large language models (LLMs), a new potential risk to the peer review process is that negligent reviewers will rely on LLMs to perform the often time consuming process of reviewing a paper. In this study, we investigate the ability of existing AI text detection algorithms to distinguish between peer reviews written by humans and different state-of-the-art LLMs. Our analysis shows that existing approaches fail to identify many GPT-4o written reviews without also producing a high number of false positive classifications. To address this deficiency, we propose a new detection approach which surpasses existing methods in the identification of GPT-4o written peer reviews at low levels of false positive classifications. Our work reveals the difficulty of accurately identifying AI-generated text at the individual review level, highlighting the urgent need for new tools and methods to detect this type of unethical application of generative AI.<|reference_end|> | arxiv | @article{yu2024is,
title={Is Your Paper Being Reviewed by an LLM? Investigating AI Text
Detectability in Peer Review},
author={Sungduk Yu, Man Luo, Avinash Madasu, Vasudev Lal, Phillip Howard},
journal={arXiv preprint arXiv:2410.03019},
year={2024},
archivePrefix={arXiv},
eprint={2410.03019},
primaryClass={cs.CL cs.AI}
} | yu2024is |
arxiv-665372 | 2410.03020 | On Logical Extrapolation for Mazes with Recurrent and Implicit Networks | <|reference_start|>On Logical Extrapolation for Mazes with Recurrent and Implicit Networks: Recent work has suggested that certain neural network architectures-particularly recurrent neural networks (RNNs) and implicit neural networks (INNs) are capable of logical extrapolation. That is, one may train such a network on easy instances of a specific task and then apply it successfully to more difficult instances of the same task. In this paper, we revisit this idea and show that (i) The capacity for extrapolation is less robust than previously suggested. Specifically, in the context of a maze-solving task, we show that while INNs (and some RNNs) are capable of generalizing to larger maze instances, they fail to generalize along axes of difficulty other than maze size. (ii) Models that are explicitly trained to converge to a fixed point (e.g. the INN we test) are likely to do so when extrapolating, while models that are not (e.g. the RNN we test) may exhibit more exotic limiting behaviour such as limit cycles, even when they correctly solve the problem. Our results suggest that (i) further study into why such networks extrapolate easily along certain axes of difficulty yet struggle with others is necessary, and (ii) analyzing the dynamics of extrapolation may yield insights into designing more efficient and interpretable logical extrapolators.<|reference_end|> | arxiv | @article{knutson2024on,
title={On Logical Extrapolation for Mazes with Recurrent and Implicit Networks},
author={Brandon Knutson and Amandin Chyba Rabeendran and Michael Ivanitskiy
and Jordan Pettyjohn and Cecilia Diniz-Behn and Samy Wu Fung and Daniel
McKenzie},
journal={arXiv preprint arXiv:2410.03020},
year={2024},
archivePrefix={arXiv},
eprint={2410.03020},
primaryClass={cs.LG stat.ML}
} | knutson2024on |
arxiv-665373 | 2410.03021 | PixelShuffler: A Simple Image Translation Through Pixel Rearrangement | <|reference_start|>PixelShuffler: A Simple Image Translation Through Pixel Rearrangement: Image-to-image translation is a topic in computer vision that has a vast range of use cases ranging from medical image translation, such as converting MRI scans to CT scans or to other MRI contrasts, to image colorization, super-resolution, domain adaptation, and generating photorealistic images from sketches or semantic maps. Image style transfer is also a widely researched application of image-to-image translation, where the goal is to synthesize an image that combines the content of one image with the style of another. Existing state-of-the-art methods often rely on complex neural networks, including diffusion models and language models, to achieve high-quality style transfer, but these methods can be computationally expensive and intricate to implement. In this paper, we propose a novel pixel shuffle method that addresses the image-to-image translation problem generally with a specific demonstrative application in style transfer. The proposed method approaches style transfer by shuffling the pixels of the style image such that the mutual information between the shuffled image and the content image is maximized. This approach inherently preserves the colors of the style image while ensuring that the structural details of the content image are retained in the stylized output. We demonstrate that this simple and straightforward method produces results that are comparable to state-of-the-art techniques, as measured by the Learned Perceptual Image Patch Similarity (LPIPS) loss for content preservation and the Fr\'echet Inception Distance (FID) score for style similarity. Our experiments validate that the proposed pixel shuffle method achieves competitive performance with significantly reduced complexity, offering a promising alternative for efficient image style transfer, as well as a promise in usability of the method in general image-to-image translation tasks.<|reference_end|> | arxiv | @article{zamzam2024pixelshuffler:,
title={PixelShuffler: A Simple Image Translation Through Pixel Rearrangement},
author={Omar Zamzam},
journal={arXiv preprint arXiv:2410.03021},
year={2024},
archivePrefix={arXiv},
eprint={2410.03021},
primaryClass={cs.CV eess.IV}
} | zamzam2024pixelshuffler: |
arxiv-665374 | 2410.03022 | Uncovering the New Accessibility Crisis in Scholarly PDFs | <|reference_start|>Uncovering the New Accessibility Crisis in Scholarly PDFs: Most scholarly works are distributed online in PDF format, which can present significant accessibility challenges for blind and low-vision readers. To characterize the scope of this issue, we perform a large-scale analysis of 20K open- and closed-access scholarly PDFs published between 2014-2023 sampled across broad fields of study. We assess the accessibility compliance of these documents based on six criteria: Default Language, Appropriate Nesting, Tagged PDF, Table Headers, Tab Order, and Alt-Text; selected based on prior work and the SIGACCESS Guide for Accessible PDFs. To ensure robustness, we corroborate our findings through automated accessibility checking, manual evaluation of alt text, comparative assessments with an alternate accessibility checker, and manual assessments with screen readers. Our findings reveal that less than 3.2% of tested PDFs satisfy all criteria, while a large majority (74.9%) fail to meet any criteria at all. Worse yet, we observe a concerning drop in PDF accessibility since 2019, largely among open-access papers, suggesting that efforts to improve document accessibility have not taken hold and are on a backslide. While investigating factors contributing to this drop, we identify key associations between fields of study, creation platforms used, models of publishing, and PDF accessibility compliance, suggesting that publisher and author choices significantly influence document accessibility. This paper highlights a new crisis in scholarly document accessibility and the need for a multi-faceted approach to address the problem, involving the development of better tools, enhanced author education, and systemic changes in academic publishing practices.<|reference_end|> | arxiv | @article{kumar2024uncovering,
title={Uncovering the New Accessibility Crisis in Scholarly PDFs},
author={Anukriti Kumar and Lucy Lu Wang},
journal={arXiv preprint arXiv:2410.03022},
year={2024},
archivePrefix={arXiv},
eprint={2410.03022},
primaryClass={cs.DL cs.HC}
} | kumar2024uncovering |
arxiv-665375 | 2410.03024 | Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting | <|reference_start|>Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting: Recent advancements in generative modeling, particularly diffusion models, have opened new directions for time series modeling, achieving state-of-the-art performance in forecasting and synthesis. However, the reliance of diffusion-based models on a simple, fixed prior complicates the generative process since the data and prior distributions differ significantly. We introduce TSFlow, a conditional flow matching (CFM) model for time series that simplifies the generative problem by combining Gaussian processes, optimal transport paths, and data-dependent prior distributions. By incorporating (conditional) Gaussian processes, TSFlow aligns the prior distribution more closely with the temporal structure of the data, enhancing both unconditional and conditional generation. Furthermore, we propose conditional prior sampling to enable probabilistic forecasting with an unconditionally trained model. In our experimental evaluation on eight real-world datasets, we demonstrate the generative capabilities of TSFlow, producing high-quality unconditional samples. Finally, we show that both conditionally and unconditionally trained models achieve competitive results in forecasting benchmarks, surpassing other methods on 6 out of 8 datasets.<|reference_end|> | arxiv | @article{kollovieh2024flow,
title={Flow Matching with Gaussian Process Priors for Probabilistic Time Series
Forecasting},
author={Marcel Kollovieh, Marten Lienen, David L"udke, Leo Schwinn, Stephan
G"unnemann},
journal={arXiv preprint arXiv:2410.03024},
year={2024},
archivePrefix={arXiv},
eprint={2410.03024},
primaryClass={cs.LG cs.AI stat.ML}
} | kollovieh2024flow |
arxiv-665376 | 2410.03026 | Characterizing Context Influence and Hallucination in Summarization | <|reference_start|>Characterizing Context Influence and Hallucination in Summarization: Although Large Language Models (LLMs) have achieved remarkable performance in numerous downstream tasks, their ubiquity has raised two significant concerns. One is that LLMs can hallucinate by generating content that contradicts relevant contextual information; the other is that LLMs can inadvertently leak private information due to input regurgitation. Many prior works have extensively studied each concern independently, but none have investigated them simultaneously. Furthermore, auditing the influence of provided context during open-ended generation with a privacy emphasis is understudied. To this end, we comprehensively characterize the influence and hallucination of contextual information during summarization. We introduce a definition for context influence and Context-Influence Decoding (CID), and then we show that amplifying the context (by factoring out prior knowledge) and the context being out of distribution with respect to prior knowledge increases the context's influence on an LLM. Moreover, we show that context influence gives a lower bound of the private information leakage of CID. We corroborate our analytical findings with experimental evaluations that show improving the F1 ROGUE-L score on CNN-DM for LLaMA 3 by $\textbf{10}$% over regular decoding also leads to $\textbf{1.5x}$ more influence by the context. Moreover, we empirically evaluate how context influence and hallucination are affected by (1) model capacity, (2) context size, (3) the length of the current response, and (4) different token $n$-grams of the context. Our code can be accessed here: https://github.com/james-flemings/context_influence.<|reference_end|> | arxiv | @article{flemings2024characterizing,
title={Characterizing Context Influence and Hallucination in Summarization},
author={James Flemings, Wanrong Zhang, Bo Jiang, Zafar Takhirov, Murali
Annavaram},
journal={arXiv preprint arXiv:2410.03026},
year={2024},
archivePrefix={arXiv},
eprint={2410.03026},
primaryClass={cs.CL cs.LG}
} | flemings2024characterizing |
arxiv-665377 | 2410.03027 | MLP-KAN: Unifying Deep Representation and Function Learning | <|reference_start|>MLP-KAN: Unifying Deep Representation and Function Learning: Recent advancements in both representation learning and function learning have demonstrated substantial promise across diverse domains of artificial intelligence. However, the effective integration of these paradigms poses a significant challenge, particularly in cases where users must manually decide whether to apply a representation learning or function learning model based on dataset characteristics. To address this issue, we introduce MLP-KAN, a unified method designed to eliminate the need for manual model selection. By integrating Multi-Layer Perceptrons (MLPs) for representation learning and Kolmogorov-Arnold Networks (KANs) for function learning within a Mixture-of-Experts (MoE) architecture, MLP-KAN dynamically adapts to the specific characteristics of the task at hand, ensuring optimal performance. Embedded within a transformer-based framework, our work achieves remarkable results on four widely-used datasets across diverse domains. Extensive experimental evaluation demonstrates its superior versatility, delivering competitive performance across both deep representation and function learning tasks. These findings highlight the potential of MLP-KAN to simplify the model selection process, offering a comprehensive, adaptable solution across various domains. Our code and weights are available at \url{https://github.com/DLYuanGod/MLP-KAN}.<|reference_end|> | arxiv | @article{he2024mlp-kan:,
title={MLP-KAN: Unifying Deep Representation and Function Learning},
author={Yunhong He, Yifeng Xie, Zhengqing Yuan, Lichao Sun},
journal={arXiv preprint arXiv:2410.03027},
year={2024},
archivePrefix={arXiv},
eprint={2410.03027},
primaryClass={cs.LG cs.CL}
} | he2024mlp-kan: |
arxiv-665378 | 2410.03030 | Dynamic Sparse Training versus Dense Training: The Unexpected Winner in Image Corruption Robustness | <|reference_start|>Dynamic Sparse Training versus Dense Training: The Unexpected Winner in Image Corruption Robustness: It is generally perceived that Dynamic Sparse Training opens the door to a new era of scalability and efficiency for artificial neural networks at, perhaps, some costs in accuracy performance for the classification task. At the same time, Dense Training is widely accepted as being the "de facto" approach to train artificial neural networks if one would like to maximize their robustness against image corruption. In this paper, we question this general practice. Consequently, we claim that, contrary to what is commonly thought, the Dynamic Sparse Training methods can consistently outperform Dense Training in terms of robustness accuracy, particularly if the efficiency aspect is not considered as a main objective (i.e., sparsity levels between 10% and up to 50%), without adding (or even reducing) resource cost. We validate our claim on two types of data, images and videos, using several traditional and modern deep learning architectures for computer vision and three widely studied Dynamic Sparse Training algorithms. Our findings reveal a new yet-unknown benefit of Dynamic Sparse Training and open new possibilities in improving deep learning robustness beyond the current state of the art.<|reference_end|> | arxiv | @article{wu2024dynamic,
title={Dynamic Sparse Training versus Dense Training: The Unexpected Winner in
Image Corruption Robustness},
author={Boqian Wu, Qiao Xiao, Shunxin Wang, Nicola Strisciuglio, Mykola
Pechenizkiy, Maurice van Keulen, Decebal Constantin Mocanu, Elena Mocanu},
journal={arXiv preprint arXiv:2410.03030},
year={2024},
archivePrefix={arXiv},
eprint={2410.03030},
primaryClass={cs.CV cs.AI}
} | wu2024dynamic |
arxiv-665379 | 2410.03031 | Single-Shot 6DoF Pose and 3D Size Estimation for Robotic Strawberry Harvesting | <|reference_start|>Single-Shot 6DoF Pose and 3D Size Estimation for Robotic Strawberry Harvesting: In this study, we introduce a deep-learning approach for determining both the 6DoF pose and 3D size of strawberries, aiming to significantly augment robotic harvesting efficiency. Our model was trained on a synthetic strawberry dataset, which is automatically generated within the Ignition Gazebo simulator, with a specific focus on the inherent symmetry exhibited by strawberries. By leveraging domain randomization techniques, the model demonstrated exceptional performance, achieving an 84.77\% average precision (AP) of 3D Intersection over Union (IoU) scores on the simulated dataset. Empirical evaluations, conducted by testing our model on real-world datasets, underscored the model's viability for real-world strawberry harvesting scenarios, even though its training was based on synthetic data. The model also exhibited robust occlusion handling abilities, maintaining accurate detection capabilities even when strawberries were obscured by other strawberries or foliage. Additionally, the model showcased remarkably swift inference speeds, reaching up to 60 frames per second (FPS).<|reference_end|> | arxiv | @article{li2024single-shot,
title={Single-Shot 6DoF Pose and 3D Size Estimation for Robotic Strawberry
Harvesting},
author={Lun Li, Hamidreza Kasaei},
journal={arXiv preprint arXiv:2410.03031},
year={2024},
archivePrefix={arXiv},
eprint={2410.03031},
primaryClass={cs.RO}
} | li2024single-shot |
arxiv-665380 | 2410.03032 | CounterQuill: Investigating the Potential of Human-AI Collaboration in Online Counterspeech Writing | <|reference_start|>CounterQuill: Investigating the Potential of Human-AI Collaboration in Online Counterspeech Writing: Online hate speech has become increasingly prevalent on social media platforms, causing harm to individuals and society. While efforts have been made to combat this issue through content moderation, the potential of user-driven counterspeech as an alternative solution remains underexplored. Existing counterspeech methods often face challenges such as fear of retaliation and skill-related barriers. To address these challenges, we introduce CounterQuill, an AI-mediated system that assists users in composing effective and empathetic counterspeech. CounterQuill provides a three-step process: (1) a learning session to help users understand hate speech and counterspeech; (2) a brainstorming session that guides users in identifying key elements of hate speech and exploring counterspeech strategies; and (3) a co-writing session that enables users to draft and refine their counterspeech with CounterQuill. We conducted a within-subjects user study with 20 participants to evaluate CounterQuill in comparison to ChatGPT. Results show that CounterQuill's guidance and collaborative writing process provided users a stronger sense of ownership over their co-authored counterspeech. Users perceived CounterQuill as a writing partner and thus were more willing to post the co-written counterspeech online compared to the one written with ChatGPT.<|reference_end|> | arxiv | @article{ding2024counterquill:,
title={CounterQuill: Investigating the Potential of Human-AI Collaboration in
Online Counterspeech Writing},
author={Xiaohan Ding, Kaike Ping, Uma Sushmitha Gunturi, Buse Carik, Sophia
Stil, Lance T Wilhelm, Taufiq Daryanto, James Hawdon, Sang Won Lee, Eugenia H
Rho},
journal={arXiv preprint arXiv:2410.03032},
year={2024},
archivePrefix={arXiv},
eprint={2410.03032},
primaryClass={cs.HC cs.AI cs.CY}
} | ding2024counterquill: |
arxiv-665381 | 2410.03035 | SPINE: Online Semantic Planning for Missions with Incomplete Natural Language Specifications in Unstructured Environments | <|reference_start|>SPINE: Online Semantic Planning for Missions with Incomplete Natural Language Specifications in Unstructured Environments: As robots become increasingly capable, users will want to describe high-level missions and have robots fill in the gaps. In many realistic settings, pre-built maps are difficult to obtain, so execution requires exploration and mapping that are necessary and specific to the mission. Consider an emergency response scenario where a user commands a robot, "triage impacted regions." The robot must infer relevant semantics (victims, etc.) and exploration targets (damaged regions) based on priors or other context, then explore and refine its plan online. These missions are incompletely specified, meaning they imply subtasks and semantics. While many semantic planning methods operate online, they are typically designed for well specified tasks such as object search or exploration. Recently, Large Language Models (LLMs) have demonstrated powerful contextual reasoning over a range of robotic tasks described in natural language. However, existing LLM planners typically do not consider online planning or complex missions; rather, relevant subtasks are provided by a pre-built map or a user. We address these limitations via SPINE (online Semantic Planner for missions with Incomplete Natural language specifications in unstructured Environments). SPINE uses an LLM to reason about subtasks implied by the mission then realizes these subtasks in a receding horizon framework. Tasks are automatically validated for safety and refined online with new observations. We evaluate SPINE in simulation and real-world settings. Evaluation missions require multiple steps of semantic reasoning and exploration in cluttered outdoor environments of over 20,000m$^2$ area. We evaluate SPINE against competitive baselines in single-agent and air-ground teaming applications. Please find videos and software on our project page: https://zacravichandran.github.io/SPINE<|reference_end|> | arxiv | @article{ravichandran2024spine:,
title={SPINE: Online Semantic Planning for Missions with Incomplete Natural
Language Specifications in Unstructured Environments},
author={Zachary Ravichandran, Varun Murali, Mariliza Tzes, George J. Pappas,
Vijay Kumar},
journal={arXiv preprint arXiv:2410.03035},
year={2024},
archivePrefix={arXiv},
eprint={2410.03035},
primaryClass={cs.RO cs.AI}
} | ravichandran2024spine: |
arxiv-665382 | 2410.03037 | Disentangling Textual and Acoustic Features of Neural Speech Representations | <|reference_start|>Disentangling Textual and Acoustic Features of Neural Speech Representations: Neural speech models build deeply entangled internal representations, which capture a variety of features (e.g., fundamental frequency, loudness, syntactic category, or semantic content of a word) in a distributed encoding. This complexity makes it difficult to track the extent to which such representations rely on textual and acoustic information, or to suppress the encoding of acoustic features that may pose privacy risks (e.g., gender or speaker identity) in critical, real-world applications. In this paper, we build upon the Information Bottleneck principle to propose a disentanglement framework that separates complex speech representations into two distinct components: one encoding content (i.e., what can be transcribed as text) and the other encoding acoustic features relevant to a given downstream task. We apply and evaluate our framework to emotion recognition and speaker identification downstream tasks, quantifying the contribution of textual and acoustic features at each model layer. Additionally, we explore the application of our disentanglement framework as an attribution method to identify the most salient speech frame representations from both the textual and acoustic perspectives.<|reference_end|> | arxiv | @article{mohebbi2024disentangling,
title={Disentangling Textual and Acoustic Features of Neural Speech
Representations},
author={Hosein Mohebbi, Grzegorz Chrupa{l}a, Willem Zuidema, Afra Alishahi,
Ivan Titov},
journal={arXiv preprint arXiv:2410.03037},
year={2024},
archivePrefix={arXiv},
eprint={2410.03037},
primaryClass={cs.CL cs.LG cs.SD eess.AS}
} | mohebbi2024disentangling |
arxiv-665383 | 2410.03038 | CPFD: Confidence-aware Privileged Feature Distillation for Short Video Classification | <|reference_start|>CPFD: Confidence-aware Privileged Feature Distillation for Short Video Classification: Dense features, customized for different business scenarios, are essential in short video classification. However, their complexity, specific adaptation requirements, and high computational costs make them resource-intensive and less accessible during online inference. Consequently, these dense features are categorized as `Privileged Dense Features'.Meanwhile, end-to-end multi-modal models have shown promising results in numerous computer vision tasks. In industrial applications, prioritizing end-to-end multi-modal features, can enhance efficiency but often leads to the loss of valuable information from historical privileged dense features. To integrate both features while maintaining efficiency and manageable resource costs, we present Confidence-aware Privileged Feature Distillation (CPFD), which empowers features of an end-to-end multi-modal model by adaptively distilling privileged features during training. Unlike existing privileged feature distillation (PFD) methods, which apply uniform weights to all instances during distillation, potentially causing unstable performance across different business scenarios and a notable performance gap between teacher model (Dense Feature enhanced multimodal-model DF-X-VLM) and student model (multimodal-model only X-VLM), our CPFD leverages confidence scores derived from the teacher model to adaptively mitigate the performance variance with the student model. We conducted extensive offline experiments on five diverse tasks demonstrating that CPFD improves the video classification F1 score by 6.76% compared with end-to-end multimodal-model (X-VLM) and by 2.31% with vanilla PFD on-average. And it reduces the performance gap by 84.6% and achieves results comparable to teacher model DF-X-VLM. The effectiveness of CPFD is further substantiated by online experiments, and our framework has been deployed in production systems for over a dozen models.<|reference_end|> | arxiv | @article{shi2024cpfd:,
title={CPFD: Confidence-aware Privileged Feature Distillation for Short Video
Classification},
author={Jinghao Shi, Xiang Shen, Kaili Zhao, Xuedong Wang, Vera Wen, Zixuan
Wang, Yifan Wu, Zhixin Zhang},
journal={arXiv preprint arXiv:2410.03038},
year={2024},
doi={10.1145/3627673.3680045},
archivePrefix={arXiv},
eprint={2410.03038},
primaryClass={cs.LG cs.CV}
} | shi2024cpfd: |
arxiv-665384 | 2410.03039 | Revealing the Unseen: Guiding Personalized Diffusion Models to Expose Training Data | <|reference_start|>Revealing the Unseen: Guiding Personalized Diffusion Models to Expose Training Data: Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot fine-tuning where a pretrained DM is fine-tuned on a small set of images to capture specific styles or objects. Many people upload these personalized checkpoints online, fostering communities such as Civitai and HuggingFace. However, model owners may overlook the potential risks of data leakage by releasing their fine-tuned checkpoints. Moreover, concerns regarding copyright violations arise when unauthorized data is used during fine-tuning. In this paper, we ask: "Can training data be extracted from these fine-tuned DMs shared online?" A successful extraction would present not only data leakage threats but also offer tangible evidence of copyright infringement. To answer this, we propose FineXtract, a framework for extracting fine-tuning data. Our method approximates fine-tuning as a gradual shift in the model's learned distribution -- from the original pretrained DM toward the fine-tuning data. By extrapolating the models before and after fine-tuning, we guide the generation toward high-probability regions within the fine-tuned data distribution. We then apply a clustering algorithm to extract the most probable images from those generated using this extrapolated guidance. Experiments on DMs fine-tuned with datasets such as WikiArt, DreamBooth, and real-world checkpoints posted online validate the effectiveness of our method, extracting approximately 20% of fine-tuning data in most cases, significantly surpassing baseline performance.<|reference_end|> | arxiv | @article{wu2024revealing,
title={Revealing the Unseen: Guiding Personalized Diffusion Models to Expose
Training Data},
author={Xiaoyu Wu, Jiaru Zhang, Steven Wu},
journal={arXiv preprint arXiv:2410.03039},
year={2024},
archivePrefix={arXiv},
eprint={2410.03039},
primaryClass={cs.CV cs.AI cs.LG}
} | wu2024revealing |
arxiv-665385 | 2410.03040 | Geometry is All You Need: A Unified Taxonomy of Matrix and Tensor Factorization for Compression of Generative Language Models | <|reference_start|>Geometry is All You Need: A Unified Taxonomy of Matrix and Tensor Factorization for Compression of Generative Language Models: Matrix and tensor-guided parametrization for Natural Language Processing (NLP) models is fundamentally useful for the improvement of the model's systematic efficiency. However, the internal links between these two algebra structures and language model parametrization are poorly understood. Also, the existing matrix and tensor research is math-heavy and far away from machine learning (ML) and NLP research concepts. These two issues result in the recent progress on matrices and tensors for model parametrization being more like a loose collection of separate components from matrix/tensor and NLP studies, rather than a well-structured unified approach, further hindering algorithm design. To this end, we propose a unified taxonomy, which bridges the matrix/tensor compression approaches and model compression concepts in ML and NLP research. Namely, we adopt an elementary concept in linear algebra, that of a subspace, which is also the core concept in geometric algebra, to reformulate the matrix/tensor and ML/NLP concepts (e.g. attention mechanism) under one umbrella. In this way, based on our subspace formalization, typical matrix and tensor decomposition algorithms can be interpreted as geometric transformations. Finally, we revisit recent literature on matrix- or tensor-guided language model compression, rephrase and compare their core ideas, and then point out the current research gap and potential solutions.<|reference_end|> | arxiv | @article{xu2024geometry,
title={Geometry is All You Need: A Unified Taxonomy of Matrix and Tensor
Factorization for Compression of Generative Language Models},
author={Mingxue Xu, Sadia Sharmin, Danilo P. Mandic},
journal={arXiv preprint arXiv:2410.03040},
year={2024},
archivePrefix={arXiv},
eprint={2410.03040},
primaryClass={cs.CL cs.LG cs.NA math.NA}
} | xu2024geometry |
arxiv-665386 | 2410.03041 | Minmax Trend Filtering: A Locally Adaptive Nonparametric Regression Method via Pointwise Min Max Optimization | <|reference_start|>Minmax Trend Filtering: A Locally Adaptive Nonparametric Regression Method via Pointwise Min Max Optimization: Trend Filtering is a nonparametric regression method which exhibits local adaptivity, in contrast to a host of classical linear smoothing methods. However, there seems to be no unanimously agreed upon definition of local adaptivity in the literature. A question we seek to answer here is how exactly is Fused Lasso or Total Variation Denoising, which is Trend Filtering of order $0$, locally adaptive? To answer this question, we first derive a new pointwise formula for the Fused Lasso estimator in terms of min-max/max-min optimization of penalized local averages. This pointwise representation appears to be new and gives a concrete explanation of the local adaptivity of Fused Lasso. It yields that the estimation error of Fused Lasso at any given point is bounded by the best (local) bias variance tradeoff where bias and variance have a slightly different meaning than usual. We then propose higher order polynomial versions of Fused Lasso which are defined pointwise in terms of min-max/max-min optimization of penalized local polynomial regressions. These appear to be new nonparametric regression methods, different from any existing method in the nonparametric regression toolbox. We call these estimators Minmax Trend Filtering. They continue to enjoy the notion of local adaptivity in the sense that their estimation error at any given point is bounded by the best (local) bias variance tradeoff.<|reference_end|> | arxiv | @article{chatterjee2024minmax,
title={Minmax Trend Filtering: A Locally Adaptive Nonparametric Regression
Method via Pointwise Min Max Optimization},
author={Sabyasachi Chatterjee},
journal={arXiv preprint arXiv:2410.03041},
year={2024},
archivePrefix={arXiv},
eprint={2410.03041},
primaryClass={math.ST cs.LG stat.TH}
} | chatterjee2024minmax |
arxiv-665387 | 2410.03042 | FedPeWS: Personalized Warmup via Subnetworks for Enhanced Heterogeneous Federated Learning | <|reference_start|>FedPeWS: Personalized Warmup via Subnetworks for Enhanced Heterogeneous Federated Learning: Statistical data heterogeneity is a significant barrier to convergence in federated learning (FL). While prior work has advanced heterogeneous FL through better optimization objectives, these methods fall short when there is extreme data heterogeneity among collaborating participants. We hypothesize that convergence under extreme data heterogeneity is primarily hindered due to the aggregation of conflicting updates from the participants in the initial collaboration rounds. To overcome this problem, we propose a warmup phase where each participant learns a personalized mask and updates only a subnetwork of the full model. This personalized warmup allows the participants to focus initially on learning specific subnetworks tailored to the heterogeneity of their data. After the warmup phase, the participants revert to standard federated optimization, where all parameters are communicated. We empirically demonstrate that the proposed personalized warmup via subnetworks (FedPeWS) approach improves accuracy and convergence speed over standard federated optimization methods.<|reference_end|> | arxiv | @article{tastan2024fedpews:,
title={FedPeWS: Personalized Warmup via Subnetworks for Enhanced Heterogeneous
Federated Learning},
author={Nurbek Tastan, Samuel Horvath, Martin Takac, Karthik Nandakumar},
journal={arXiv preprint arXiv:2410.03042},
year={2024},
archivePrefix={arXiv},
eprint={2410.03042},
primaryClass={cs.LG cs.DC}
} | tastan2024fedpews: |
arxiv-665388 | 2410.03043 | Towards Understanding the Feasibility of Machine Unlearning | <|reference_start|>Towards Understanding the Feasibility of Machine Unlearning: In light of recent privacy regulations, machine unlearning has attracted significant attention in the research community. However, current studies predominantly assess the overall success of unlearning approaches, overlooking the varying difficulty of unlearning individual training samples. As a result, the broader feasibility of machine unlearning remains under-explored. This paper presents a set of novel metrics for quantifying the difficulty of unlearning by jointly considering the properties of target model and data distribution. Specifically, we propose several heuristics to assess the conditions necessary for a successful unlearning operation, examine the variations in unlearning difficulty across different training samples, and present a ranking mechanism to identify the most challenging samples to unlearn. We highlight the effectiveness of the Kernelized Stein Discrepancy (KSD), a parameterized kernel function tailored to each model and dataset, as a heuristic for evaluating unlearning difficulty. Our approach is validated through multiple classification tasks and established machine unlearning algorithms, demonstrating the practical feasibility of unlearning operations across diverse scenarios.<|reference_end|> | arxiv | @article{sarvmaili2024towards,
title={Towards Understanding the Feasibility of Machine Unlearning},
author={Mahtab Sarvmaili, Hassan Sajjad, Ga Wu},
journal={arXiv preprint arXiv:2410.03043},
year={2024},
archivePrefix={arXiv},
eprint={2410.03043},
primaryClass={cs.LG}
} | sarvmaili2024towards |
arxiv-665389 | 2410.03045 | Vehicle Suspension Recommendation System: Multi-Fidelity Neural Network-based Mechanism Design Optimization | <|reference_start|>Vehicle Suspension Recommendation System: Multi-Fidelity Neural Network-based Mechanism Design Optimization: Mechanisms are designed to perform functions in various fields. Often, there is no unique mechanism that performs a well-defined function. For example, vehicle suspensions are designed to improve driving performance and ride comfort, but different types are available depending on the environment. This variability in design makes performance comparison difficult. Additionally, the traditional design process is multi-step, gradually reducing the number of design candidates while performing costly analyses to meet target performance. Recently, AI models have been used to reduce the computational cost of FEA. However, there are limitations in data availability and different analysis environments, especially when transitioning from low-fidelity to high-fidelity analysis. In this paper, we propose a multi-fidelity design framework aimed at recommending optimal types and designs of mechanical mechanisms. As an application, vehicle suspension systems were selected, and several types were defined. For each type, mechanism parameters were generated and converted into 3D CAD models, followed by low-fidelity rigid body dynamic analysis under driving conditions. To effectively build a deep learning-based multi-fidelity surrogate model, the results of the low-fidelity analysis were analyzed using DBSCAN and sampled at 5% for high-cost flexible body dynamic analysis. After training the multi-fidelity model, a multi-objective optimization problem was formulated for the performance metrics of each suspension type. Finally, we recommend the optimal type and design based on the input to optimize ride comfort-related performance metrics. To validate the proposed methodology, we extracted basic design rules of Pareto solutions using data mining techniques. We also verified the effectiveness and applicability by comparing the results with those obtained from a conventional deep learning-based design process.<|reference_end|> | arxiv | @article{lee2024vehicle,
title={Vehicle Suspension Recommendation System: Multi-Fidelity Neural
Network-based Mechanism Design Optimization},
author={Sumin Lee and Namwoo Kang},
journal={arXiv preprint arXiv:2410.03045},
year={2024},
archivePrefix={arXiv},
eprint={2410.03045},
primaryClass={physics.comp-ph cs.LG}
} | lee2024vehicle |
arxiv-665390 | 2410.03049 | Scalable Frame-based Construction of Sociocultural NormBases for Socially-Aware Dialogues | <|reference_start|>Scalable Frame-based Construction of Sociocultural NormBases for Socially-Aware Dialogues: Sociocultural norms serve as guiding principles for personal conduct in social interactions, emphasizing respect, cooperation, and appropriate behavior, which is able to benefit tasks including conversational information retrieval, contextual information retrieval and retrieval-enhanced machine learning. We propose a scalable approach for constructing a Sociocultural Norm (SCN) Base using Large Language Models (LLMs) for socially aware dialogues. We construct a comprehensive and publicly accessible Chinese Sociocultural NormBase. Our approach utilizes socially aware dialogues, enriched with contextual frames, as the primary data source to constrain the generating process and reduce the hallucinations. This enables extracting of high-quality and nuanced natural-language norm statements, leveraging the pragmatic implications of utterances with respect to the situation. As real dialogue annotated with gold frames are not readily available, we propose using synthetic data. Our empirical results show: (i) the quality of the SCNs derived from synthetic data is comparable to that from real dialogues annotated with gold frames, and (ii) the quality of the SCNs extracted from real data, annotated with either silver (predicted) or gold frames, surpasses that without the frame annotations. We further show the effectiveness of the extracted SCNs in a RAG-based (Retrieval-Augmented Generation) model to reason about multiple downstream dialogue tasks.<|reference_end|> | arxiv | @article{qu2024scalable,
title={Scalable Frame-based Construction of Sociocultural NormBases for
Socially-Aware Dialogues},
author={Shilin Qu, Weiqing Wang, Xin Zhou, Haolan Zhan, Zhuang Li, Lizhen Qu,
Linhao Luo, Yuan-Fang Li, and Gholamreza Haffari},
journal={TOMM 2024},
year={2024},
archivePrefix={arXiv},
eprint={2410.03049},
primaryClass={cs.CL cs.AI cs.IR cs.LG}
} | qu2024scalable |
arxiv-665391 | 2410.03051 | AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark | <|reference_start|>AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark: Video detailed captioning is a key task which aims to generate comprehensive and coherent textual descriptions of video content, benefiting both video understanding and generation. In this paper, we propose AuroraCap, a video captioner based on a large multimodal model. We follow the simplest architecture design without additional parameters for temporal modeling. To address the overhead caused by lengthy video sequences, we implement the token merging strategy, reducing the number of input visual tokens. Surprisingly, we found that this strategy results in little performance loss. AuroraCap shows superior performance on various video and image captioning benchmarks, for example, obtaining a CIDEr of 88.9 on Flickr30k, beating GPT-4V (55.3) and Gemini-1.5 Pro (82.2). However, existing video caption benchmarks only include simple descriptions, consisting of a few dozen words, which limits research in this field. Therefore, we develop VDC, a video detailed captioning benchmark with over one thousand carefully annotated structured captions. In addition, we propose a new LLM-assisted metric VDCscore for bettering evaluation, which adopts a divide-and-conquer strategy to transform long caption evaluation into multiple short question-answer pairs. With the help of human Elo ranking, our experiments show that this benchmark better correlates with human judgments of video detailed captioning quality.<|reference_end|> | arxiv | @article{chai2024auroracap:,
title={AuroraCap: Efficient, Performant Video Detailed Captioning and a New
Benchmark},
author={Wenhao Chai, Enxin Song, Yilun Du, Chenlin Meng, Vashisht Madhavan,
Omer Bar-Tal, Jeng-Neng Hwang, Saining Xie, Christopher D. Manning},
journal={arXiv preprint arXiv:2410.03051},
year={2024},
archivePrefix={arXiv},
eprint={2410.03051},
primaryClass={cs.CV}
} | chai2024auroracap: |
arxiv-665392 | 2410.03052 | Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport | <|reference_start|>Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport: To embed structured knowledge within labels into feature representations, prior work (Zeng et al., 2022) proposed to use the Cophenetic Correlation Coefficient (CPCC) as a regularizer during supervised learning. This regularizer calculates pairwise Euclidean distances of class means and aligns them with the corresponding shortest path distances derived from the label hierarchy tree. However, class means may not be good representatives of the class conditional distributions, especially when they are multi-mode in nature. To address this limitation, under the CPCC framework, we propose to use the Earth Mover's Distance (EMD) to measure the pairwise distances among classes in the feature space. We show that our exact EMD method generalizes previous work, and recovers the existing algorithm when class-conditional distributions are Gaussian in the feature space. To further improve the computational efficiency of our method, we introduce the Optimal Transport-CPCC family by exploring four EMD approximation variants. Our most efficient OT-CPCC variant runs in linear time in the size of the dataset, while maintaining competitive performance across datasets and tasks.<|reference_end|> | arxiv | @article{zeng2024learning,
title={Learning Structured Representations by Embedding Class Hierarchy with
Fast Optimal Transport},
author={Siqi Zeng, Sixian Du, Makoto Yamada, Han Zhao},
journal={arXiv preprint arXiv:2410.03052},
year={2024},
archivePrefix={arXiv},
eprint={2410.03052},
primaryClass={cs.LG}
} | zeng2024learning |
arxiv-665393 | 2410.03054 | CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision Language Models for Object-based Global Localization | <|reference_start|>CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision Language Models for Object-based Global Localization: This letter proposes a method of global localization on a map with semantic object landmarks. One of the most promising approaches for localization on object maps is to use semantic graph matching using landmark descriptors calculated from the distribution of surrounding objects. These descriptors are vulnerable to misclassification and partial observations. Moreover, many existing methods rely on inlier extraction using RANSAC, which is stochastic and sensitive to a high outlier rate. To address the former issue, we augment the correspondence matching using Vision Language Models (VLMs). Landmark discriminability is improved by VLM embeddings, which are independent of surrounding objects. In addition, inliers are estimated deterministically using a graph-theoretic approach. We also incorporate pose calculation using the weighted least squares considering correspondence similarity and observation completeness to improve the robustness. We confirmed improvements in matching and pose estimation accuracy through experiments on ScanNet and TUM datasets.<|reference_end|> | arxiv | @article{matsuzaki2024clip-clique:,
title={CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision
Language Models for Object-based Global Localization},
author={Shigemichi Matsuzaki, Kazuhito Tanaka, Kazuhiro Shintani},
journal={arXiv preprint arXiv:2410.03054},
year={2024},
doi={10.1109/LRA.2024.3474482},
archivePrefix={arXiv},
eprint={2410.03054},
primaryClass={cs.CV cs.RO}
} | matsuzaki2024clip-clique: |
arxiv-665394 | 2410.03055 | Permissive Information-Flow Analysis for Large Language Models | <|reference_start|>Permissive Information-Flow Analysis for Large Language Models: Large Language Models (LLMs) are rapidly becoming commodity components of larger software systems. This poses natural security and privacy problems: poisoned data retrieved from one component can change the model's behavior and compromise the entire system, including coercing the model to spread confidential data to untrusted components. One promising approach is to tackle this problem at the system level via dynamic information flow (aka taint) tracking. Unfortunately, the traditional approach of propagating the most restrictive input label to the output is too conservative for applications where LLMs operate on inputs retrieved from diverse sources. In this paper, we propose a novel, more permissive approach to propagate information flow labels through LLM queries. The key idea behind our approach is to propagate only the labels of the samples that were influential in generating the model output and to eliminate the labels of unnecessary input. We implement and investigate the effectiveness of two variations of this approach, based on (i) prompt-based retrieval augmentation, and (ii) a $k$-nearest-neighbors language model. We compare these with the baseline of an introspection-based influence estimator that directly asks the language model to predict the output label. The results obtained highlight the superiority of our prompt-based label propagator, which improves the label in more than 85% of the cases in an LLM agent setting. These findings underscore the practicality of permissive label propagation for retrieval augmentation.<|reference_end|> | arxiv | @article{siddiqui2024permissive,
title={Permissive Information-Flow Analysis for Large Language Models},
author={Shoaib Ahmed Siddiqui, Radhika Gaonkar, Boris K"opf, David Krueger,
Andrew Paverd, Ahmed Salem, Shruti Tople, Lukas Wutschitz, Menglin Xia,
Santiago Zanella-B'eguelin},
journal={arXiv preprint arXiv:2410.03055},
year={2024},
archivePrefix={arXiv},
eprint={2410.03055},
primaryClass={cs.LG cs.AI}
} | siddiqui2024permissive |
arxiv-665395 | 2410.03056 | Towards an Improved Metric for Evaluating Disentangled Representations | <|reference_start|>Towards an Improved Metric for Evaluating Disentangled Representations: Disentangled representation learning plays a pivotal role in making representations controllable, interpretable and transferable. Despite its significance in the domain, the quest for reliable and consistent quantitative disentanglement metric remains a major challenge. This stems from the utilisation of diverse metrics measuring different properties and the potential bias introduced by their design. Our work undertakes a comprehensive examination of existing popular disentanglement evaluation metrics, comparing them in terms of measuring aspects of disentanglement (viz. Modularity, Compactness, and Explicitness), detecting the factor-code relationship, and describing the degree of disentanglement. We propose a new framework for quantifying disentanglement, introducing a metric entitled \emph{EDI}, that leverages the intuitive concept of \emph{exclusivity} and improved factor-code relationship to minimize ad-hoc decisions. An in-depth analysis reveals that EDI measures essential properties while offering more stability than existing metrics, advocating for its adoption as a standardised approach.<|reference_end|> | arxiv | @article{julka2024towards,
title={Towards an Improved Metric for Evaluating Disentangled Representations},
author={Sahib Julka, Yashu Wang, Michael Granitzer},
journal={arXiv preprint arXiv:2410.03056},
year={2024},
archivePrefix={arXiv},
eprint={2410.03056},
primaryClass={cs.LG cs.AI}
} | julka2024towards |
arxiv-665396 | 2410.03057 | How to evaluate your medical time series classification? | <|reference_start|>How to evaluate your medical time series classification?: Medical time series (MedTS) play a critical role in many healthcare applications, such as vital sign monitoring and the diagnosis of brain and heart diseases. However, the existence of subject-specific features poses unique challenges in MedTS evaluation. Inappropriate evaluation setups that either exploit or overlook these features can lead to artificially inflated classification performance (by up to 50% in accuracy on ADFTD dataset): this concern has received little attention in current research. Here, we categorize the existing evaluation setups into two primary categories: subject-dependent and subject-independent. We show the subject-independent setup is more appropriate for different datasets and tasks. Our theoretical analysis explores the feature components of MedTS, examining how different evaluation setups influence the features that a model learns. Through experiments on six datasets (spanning EEG, ECG, and fNIRS modalities) using four different methods, we demonstrate step-by-step how subject-dependent utilizes subject-specific features as a shortcut for classification and leads to a deceptive high performance, suggesting that the subject-independent setup is more precise and practicable evaluation setup in real-world. This comprehensive analysis aims to establish clearer guidelines for evaluating MedTS models in different healthcare applications. Code to reproduce this work in \url{https://github.com/DL4mHealth/MedTS_Evaluation}.<|reference_end|> | arxiv | @article{wang2024how,
title={How to evaluate your medical time series classification?},
author={Yihe Wang, Taida Li, Yujun Yan, Wenzhan Song, Xiang Zhang},
journal={arXiv preprint arXiv:2410.03057},
year={2024},
archivePrefix={arXiv},
eprint={2410.03057},
primaryClass={cs.CE}
} | wang2024how |
arxiv-665397 | 2410.03058 | DiffKillR: Killing and Recreating Diffeomorphisms for Cell Annotation in Dense Microscopy Images | <|reference_start|>DiffKillR: Killing and Recreating Diffeomorphisms for Cell Annotation in Dense Microscopy Images: The proliferation of digital microscopy images, driven by advances in automated whole slide scanning, presents significant opportunities for biomedical research and clinical diagnostics. However, accurately annotating densely packed information in these images remains a major challenge. To address this, we introduce DiffKillR, a novel framework that reframes cell annotation as the combination of archetype matching and image registration tasks. DiffKillR employs two complementary neural networks: one that learns a diffeomorphism-invariant feature space for robust cell matching and another that computes the precise warping field between cells for annotation mapping. Using a small set of annotated archetypes, DiffKillR efficiently propagates annotations across large microscopy images, reducing the need for extensive manual labeling. More importantly, it is suitable for any type of pixel-level annotation. We will discuss the theoretical properties of DiffKillR and validate it on three microscopy tasks, demonstrating its advantages over existing supervised, semi-supervised, and unsupervised methods.<|reference_end|> | arxiv | @article{liu2024diffkillr:,
title={DiffKillR: Killing and Recreating Diffeomorphisms for Cell Annotation in
Dense Microscopy Images},
author={Chen Liu, Danqi Liao, Alejandro Parada-Mayorga, Alejandro Ribeiro,
Marcello DiStasio, Smita Krishnaswamy},
journal={arXiv preprint arXiv:2410.03058},
year={2024},
archivePrefix={arXiv},
eprint={2410.03058},
primaryClass={cs.CV}
} | liu2024diffkillr: |
arxiv-665398 | 2410.03061 | DocKD: Knowledge Distillation from LLMs for Open-World Document Understanding Models | <|reference_start|>DocKD: Knowledge Distillation from LLMs for Open-World Document Understanding Models: Visual document understanding (VDU) is a challenging task that involves understanding documents across various modalities (text and image) and layouts (forms, tables, etc.). This study aims to enhance generalizability of small VDU models by distilling knowledge from LLMs. We identify that directly prompting LLMs often fails to generate informative and useful data. In response, we present a new framework (called DocKD) that enriches the data generation process by integrating external document knowledge. Specifically, we provide an LLM with various document elements like key-value pairs, layouts, and descriptions, to elicit open-ended answers. Our experiments show that DocKD produces high-quality document annotations and surpasses the direct knowledge distillation approach that does not leverage external document knowledge. Moreover, student VDU models trained with solely DocKD-generated data are not only comparable to those trained with human-annotated data on in-domain tasks but also significantly excel them on out-of-domain tasks.<|reference_end|> | arxiv | @article{kim2024dockd:,
title={DocKD: Knowledge Distillation from LLMs for Open-World Document
Understanding Models},
author={Sungnyun Kim, Haofu Liao, Srikar Appalaraju, Peng Tang, Zhuowen Tu,
Ravi Kumar Satzoda, R. Manmatha, Vijay Mahadevan, Stefano Soatto},
journal={arXiv preprint arXiv:2410.03061},
year={2024},
archivePrefix={arXiv},
eprint={2410.03061},
primaryClass={cs.CV cs.CL}
} | kim2024dockd: |
arxiv-665399 | 2410.03062 | Image First or Text First? Optimising the Sequencing of Modalities in Large Language Model Prompting and Reasoning Tasks | <|reference_start|>Image First or Text First? Optimising the Sequencing of Modalities in Large Language Model Prompting and Reasoning Tasks: This paper examines how the sequencing of images and text within multi-modal prompts influences the reasoning performance of large language models (LLMs). We performed empirical evaluations using three commercial LLMs. Our results demonstrate that the order in which modalities are presented can significantly affect performance, particularly in tasks of varying complexity. For simpler tasks involving a single image, modality sequencing had a clear impact on accuracy. However, in more complex tasks involving multiple images and intricate reasoning steps, the effect of sequencing diminished, likely due to the increased cognitive demands of the task. Our findings also highlight the importance of question/prompt structure. In nested and multi-step reasoning tasks, modality sequencing played a key role in shaping model performance. While LLMs excelled in the initial stages of reasoning, they struggled to re-incorporate earlier information, underscoring the challenges of multi-hop reasoning within transformer architectures. This suggests that aligning the sequence of modalities with the logical flow of reasoning steps is more critical than modality order alone. These insights offer valuable implications for improving multi-modal prompt design, with broader applications across fields such as education, medical imaging, and cross-modal learning.<|reference_end|> | arxiv | @article{wardle2024image,
title={Image First or Text First? Optimising the Sequencing of Modalities in
Large Language Model Prompting and Reasoning Tasks},
author={Grant Wardle and Teo Susnjak},
journal={arXiv preprint arXiv:2410.03062},
year={2024},
archivePrefix={arXiv},
eprint={2410.03062},
primaryClass={cs.AI}
} | wardle2024image |
arxiv-665400 | 2410.03063 | Integrating Natural Language Prompting Tasks in Introductory Programming Courses | <|reference_start|>Integrating Natural Language Prompting Tasks in Introductory Programming Courses: Introductory programming courses often emphasize mastering syntax and basic constructs before progressing to more complex and interesting programs. This bottom-up approach can be frustrating for novices, shifting the focus away from problem solving and potentially making computing less appealing to a broad range of students. The rise of generative AI for code production could partially address these issues by fostering new skills via interaction with AI models, including constructing high-level prompts and evaluating code that is automatically generated. In this experience report, we explore the inclusion of two prompt-focused activities in an introductory course, implemented across four labs in a six-week module. The first requires students to solve computational problems by writing natural language prompts, emphasizing problem-solving over syntax. The second involves students crafting prompts to generate code equivalent to provided fragments, to foster an understanding of the relationship between prompts and code. Most of the students in the course had reported finding programming difficult to learn, often citing frustrations with syntax and debugging. We found that self-reported difficulty with learning programming had a strong inverse relationship with performance on traditional programming assessments such as tests and projects, as expected. However, performance on the natural language tasks was less strongly related to self-reported difficulty, suggesting they may target different skills. Learning how to communicate with AI coding models is becoming an important skill, and natural language prompting tasks may appeal to a broad range of students.<|reference_end|> | arxiv | @article{kerslake2024integrating,
title={Integrating Natural Language Prompting Tasks in Introductory Programming
Courses},
author={Chris Kerslake, Paul Denny, David H Smith IV, James Prather, Juho
Leinonen, Andrew Luxton-Reilly, Stephen MacNeil},
journal={arXiv preprint arXiv:2410.03063},
year={2024},
archivePrefix={arXiv},
eprint={2410.03063},
primaryClass={cs.CY cs.AI}
} | kerslake2024integrating |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.