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-663001
|
2409.19152
|
MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motion
|
<|reference_start|>MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motion: Structure-from-Motion (SfM), a task aiming at jointly recovering camera poses and 3D geometry of a scene given a set of images, remains a hard problem with still many open challenges despite decades of significant progress. The traditional solution for SfM consists of a complex pipeline of minimal solvers which tends to propagate errors and fails when images do not sufficiently overlap, have too little motion, etc. Recent methods have attempted to revisit this paradigm, but we empirically show that they fall short of fixing these core issues. In this paper, we propose instead to build upon a recently released foundation model for 3D vision that can robustly produce local 3D reconstructions and accurate matches. We introduce a low-memory approach to accurately align these local reconstructions in a global coordinate system. We further show that such foundation models can serve as efficient image retrievers without any overhead, reducing the overall complexity from quadratic to linear. Overall, our novel SfM pipeline is simple, scalable, fast and truly unconstrained, i.e. it can handle any collection of images, ordered or not. Extensive experiments on multiple benchmarks show that our method provides steady performance across diverse settings, especially outperforming existing methods in small- and medium-scale settings.<|reference_end|>
|
arxiv
|
@article{duisterhof2024mast3r-sfm:,
title={MASt3R-SfM: a Fully-Integrated Solution for Unconstrained
Structure-from-Motion},
author={Bardienus Duisterhof, Lojze Zust, Philippe Weinzaepfel, Vincent Leroy,
Yohann Cabon, and Jerome Revaud},
journal={arXiv preprint arXiv:2409.19152},
year={2024},
archivePrefix={arXiv},
eprint={2409.19152},
primaryClass={cs.CV}
}
|
duisterhof2024mast3r-sfm:
|
arxiv-663002
|
2409.19154
|
SAMBA: Scalable Approximate Forwarding For NDN Implicit FIB Aggregation
|
<|reference_start|>SAMBA: Scalable Approximate Forwarding For NDN Implicit FIB Aggregation: The Internet landscape has witnessed a significant shift toward Information Centric Networking (ICN) due to the exponential growth of data-driven applications. Similar to routing tables in TCP/IP architectures, ICN uses Forward Information Base (FIB) tables. However, FIB tables can grow exponentially due to their URL-like naming scheme, introducing major delays in the prefix lookup process. Existing explicit FIB aggregation solutions are very complex to run, and ICN on-demand routing schemes, which use a discovery mechanism to help reduce the number of FIB records and thus have shorter lookup times, rely on flooding-based mechanisms and building routes for all requests, introducing additional scalability challenges. In this paper, we propose SAMBA, an Approximate Forwarding-based Self Learning, that uses the nearest FIB trie record to the given prefix for reducing the number of discoveries thus keeping the FIB table small. By choosing the nearest prefix to a given name prefix, SAMBA uses Implicit Prefix Aggregation (IPA) which implicitly aggregates the FIB records and reduces the number of Self Learning discoveries required. Coupled with the approximate forwarding, SAMBA can achieve efficient and scalable forwarding<|reference_end|>
|
arxiv
|
@article{esmaeili2024samba:,
title={SAMBA: Scalable Approximate Forwarding For NDN Implicit FIB Aggregation},
author={Amir Esmaeili, Abderrahmen Mtibaa},
journal={arXiv preprint arXiv:2409.19154},
year={2024},
archivePrefix={arXiv},
eprint={2409.19154},
primaryClass={cs.NI}
}
|
esmaeili2024samba:
|
arxiv-663003
|
2409.19155
|
SensoPatch: A Reconfigurable Haptic Feedback with High-Density Tactile Sensing Glove
|
<|reference_start|>SensoPatch: A Reconfigurable Haptic Feedback with High-Density Tactile Sensing Glove: Haptic feedback is integral to the improved experience of prosthetic users and the reduction in prosthesis rejection. Prior studies have explored various methods to encode tactile information and deliver vibration feedback. However, a comprehensive study comparing performance across different stimulation locations and feedback modalities for wearable devices is absent and there is no test platform. This paper proposes an open-source reconfigurable haptic feedback system which incorporates 25 sensors and wireless communication to allow customized number of vibration motors, adjustable motor placement, and programmable encoding of tactile data to change feedback modalities. To demonstrate potential studies that can be investigated using SensoPatch, we conducted two experiments: 1) to assess the vibration discrimination accuracy on 3 body parts 2) to assess the effect of 6 methods of mapping tactile data to varying number of motors on object manipulation. SensoPatch utilizes low-cost off-the-shelf components, enabling large-scale comparative studies of feedback modalities and stimulation sites to optimize vibrotactile feedback and facilitate its deployment in upper limb prostheses.<|reference_end|>
|
arxiv
|
@article{angkanapiwat2024sensopatch:,
title={SensoPatch: A Reconfigurable Haptic Feedback with High-Density Tactile
Sensing Glove},
author={Yanisa Angkanapiwat, Ariel Slepyan, Rebecca J. Greene, Nitish Thakor},
journal={arXiv preprint arXiv:2409.19155},
year={2024},
archivePrefix={arXiv},
eprint={2409.19155},
primaryClass={cs.HC cs.SY eess.SY}
}
|
angkanapiwat2024sensopatch:
|
arxiv-663004
|
2409.19156
|
ZERNIPAX: A Fast and Accurate Zernike Polynomial Calculator in Python
|
<|reference_start|>ZERNIPAX: A Fast and Accurate Zernike Polynomial Calculator in Python: Zernike Polynomials serve as an orthogonal basis on the unit disc, and have been proven to be effective in optics simulations, astrophysics, and more recently in plasma simulations. Unlike Bessel functions, they maintain finite values at the disc center, ensuring inherent analyticity along the axis. We developed ZERNIPAX, an open-source Python package capable of utilizing CPU/GPUs, leveraging Google's JAX package and available on https://github.com/PlasmaControl/FastZernike.git as well as PyPI. Our implementation of the recursion relation between Jacobi polynomials significantly improves computation time compared to alternative methods by use of parallel computing while still preserving accuracy for mode numbers n>100.<|reference_end|>
|
arxiv
|
@article{elmacioglu2024zernipax:,
title={ZERNIPAX: A Fast and Accurate Zernike Polynomial Calculator in Python},
author={Yigit Gunsur Elmacioglu, Rory Conlin, Daniel W. Dudt, Dario Panici and
Egemen Kolemen},
journal={arXiv preprint arXiv:2409.19156},
year={2024},
archivePrefix={arXiv},
eprint={2409.19156},
primaryClass={cs.PF}
}
|
elmacioglu2024zernipax:
|
arxiv-663005
|
2409.19157
|
Calibrated Probabilistic Forecasts for Arbitrary Sequences
|
<|reference_start|>Calibrated Probabilistic Forecasts for Arbitrary Sequences: Real-world data streams can change unpredictably due to distribution shifts, feedback loops and adversarial actors, which challenges the validity of forecasts. We present a forecasting framework ensuring valid uncertainty estimates regardless of how data evolves. Leveraging the concept of Blackwell approachability from game theory, we introduce a forecasting framework that guarantees calibrated uncertainties for outcomes in any compact space (e.g., classification or bounded regression). We extend this framework to recalibrate existing forecasters, guaranteeing accurate uncertainties without sacrificing predictive performance. We implement both general-purpose gradient-based algorithms and algorithms optimized for popular special cases of our framework. Empirically, our algorithms improve calibration and downstream decision-making for energy systems.<|reference_end|>
|
arxiv
|
@article{marx2024calibrated,
title={Calibrated Probabilistic Forecasts for Arbitrary Sequences},
author={Charles Marx, Volodymyr Kuleshov, Stefano Ermon},
journal={arXiv preprint arXiv:2409.19157},
year={2024},
archivePrefix={arXiv},
eprint={2409.19157},
primaryClass={cs.LG stat.ML}
}
|
marx2024calibrated
|
arxiv-663006
|
2409.19158
|
bnRep: A repository of Bayesian networks from the academic literature
|
<|reference_start|>bnRep: A repository of Bayesian networks from the academic literature: Bayesian networks (BNs) are widely used for modeling complex systems with uncertainty, yet repositories of pre-built BNs remain limited. This paper introduces bnRep, an open-source R package offering a comprehensive collection of documented BNs, facilitating benchmarking, replicability, and education. With over 200 networks from academic publications, bnRep integrates seamlessly with bnlearn and other R packages, providing users with interactive tools for network exploration.<|reference_end|>
|
arxiv
|
@article{leonelli2024bnrep:,
title={bnRep: A repository of Bayesian networks from the academic literature},
author={Manuele Leonelli},
journal={arXiv preprint arXiv:2409.19158},
year={2024},
archivePrefix={arXiv},
eprint={2409.19158},
primaryClass={cs.AI physics.soc-ph}
}
|
leonelli2024bnrep:
|
arxiv-663007
|
2409.19160
|
Boundary Integral Formulations for Flexural Wave Scattering in Thin Plates
|
<|reference_start|>Boundary Integral Formulations for Flexural Wave Scattering in Thin Plates: In this paper we develop second kind integral formulations for flexural wave scattering problems involving the clamped, free, and supported plate boundary conditions. While the clamped plate problem can be solved with layer potentials previously developed for the biharmonic equation [1], the free plate problem is more difficult due to the complex nature of the boundary conditions. In this paper we describe a representation for the free plate problem that uses the Hilbert transform to cancel singularities of certain layer potentials, ultimately leading to a Fredholm integral equation of the second kind. Additionally, for the supported plate problem, we improve on an existing representation to obtain a second kind integral equation. With these representations, it is possible to solve flexural wave scattering problems with high-order-accurate methods, examine the far-field patterns of scattering objects, and solve large problems involving multiple scatterers.<|reference_end|>
|
arxiv
|
@article{nekrasov2024boundary,
title={Boundary Integral Formulations for Flexural Wave Scattering in Thin
Plates},
author={Peter Nekrasov, Tim Su, Travis Askham, Jeremy G. Hoskins},
journal={arXiv preprint arXiv:2409.19160},
year={2024},
archivePrefix={arXiv},
eprint={2409.19160},
primaryClass={math.NA cs.NA math-ph math.MP}
}
|
nekrasov2024boundary
|
arxiv-663008
|
2409.19166
|
Understanding #vent Channels on Discord
|
<|reference_start|>Understanding #vent Channels on Discord: Vent channels on Discord, which are chat channels developed for people to express frustrations, can become an informal type of peer support system. This paper is a qualitative study of experiences with vent channels on Discord, examining the experiences of 13 participants through semi-structured interviews. We find that participants are able to meet their needs for social support via vent channels by receiving commiseration, advice, and validation from the responses of others. At the same time, vent channels can lead to frustration when participants have conflicting expectations for their interactions. We suggest ways that Discord or Discord server moderators can provide enhanced structure, clarity, and transparency in order to enable participants to have better experiences in vent channels.<|reference_end|>
|
arxiv
|
@article{oladeji2024understanding,
title={Understanding #vent Channels on Discord},
author={Kayode Oladeji, Tony Wang, Diyi Yang, Amy Bruckman},
journal={arXiv preprint arXiv:2409.19166},
year={2024},
archivePrefix={arXiv},
eprint={2409.19166},
primaryClass={cs.SI cs.HC}
}
|
oladeji2024understanding
|
arxiv-663009
|
2409.19168
|
Optimization-based Task and Motion Planning under Signal Temporal Logic Specifications using Logic Network Flow
|
<|reference_start|>Optimization-based Task and Motion Planning under Signal Temporal Logic Specifications using Logic Network Flow: This paper proposes an optimization-based task and motion planning framework, named ``Logic Network Flow", to integrate signal temporal logic (STL) specifications into efficient mixed-binary linear programmings. In this framework, temporal predicates are encoded as polyhedron constraints on each edge of the network flow, instead of as constraints between the nodes as in the traditional Logic Tree formulation. Synthesized with Dynamic Network Flows, Logic Network Flows render a tighter convex relaxation compared to Logic Trees derived from these STL specifications. Our formulation is evaluated on several multi-robot motion planning case studies. Empirical results demonstrate that our formulation outperforms Logic Tree formulation in terms of computation time for several planning problems. As the problem size scales up, our method still discovers better lower and upper bounds by exploring fewer number of nodes during the branch-and-bound process, although this comes at the cost of increased computational load for each node when exploring branches.<|reference_end|>
|
arxiv
|
@article{lin2024optimization-based,
title={Optimization-based Task and Motion Planning under Signal Temporal Logic
Specifications using Logic Network Flow},
author={Xuan Lin, Jiming Ren, Samuel Coogan, and Ye Zhao},
journal={arXiv preprint arXiv:2409.19168},
year={2024},
archivePrefix={arXiv},
eprint={2409.19168},
primaryClass={cs.RO cs.FL}
}
|
lin2024optimization-based
|
arxiv-663010
|
2409.19169
|
TwinCL: A Twin Graph Contrastive Learning Model for Collaborative Filtering
|
<|reference_start|>TwinCL: A Twin Graph Contrastive Learning Model for Collaborative Filtering: In the domain of recommendation and collaborative filtering, Graph Contrastive Learning (GCL) has become an influential approach. Nevertheless, the reasons for the effectiveness of contrastive learning are still not well understood. In this paper, we challenge the conventional use of random augmentations on graph structure or embedding space in GCL, which may disrupt the structural and semantic information inherent in Graph Neural Networks. Moreover, fixed-rate data augmentation proves to be less effective compared to augmentation with an adaptive rate. In the initial training phases, significant perturbations are more suitable, while as the training approaches convergence, milder perturbations yield better results. We introduce a twin encoder in place of random augmentations, demonstrating the redundancy of traditional augmentation techniques. The twin encoder updating mechanism ensures the generation of more diverse contrastive views in the early stages, transitioning to views with greater similarity as training progresses. In addition, we investigate the learned representations from the perspective of alignment and uniformity on a hypersphere to optimize more efficiently. Our proposed Twin Graph Contrastive Learning model -- TwinCL -- aligns positive pairs of user and item embeddings and the representations from the twin encoder while maintaining the uniformity of the embeddings on the hypersphere. Our theoretical analysis and experimental results show that the proposed model optimizing alignment and uniformity with the twin encoder contributes to better recommendation accuracy and training efficiency performance. In comprehensive experiments on three public datasets, our proposed TwinCL achieves an average improvement of 5.6% (NDCG@10) in recommendation accuracy with faster training speed, while effectively mitigating popularity bias.<|reference_end|>
|
arxiv
|
@article{liu2024twincl:,
title={TwinCL: A Twin Graph Contrastive Learning Model for Collaborative
Filtering},
author={Chengkai Liu, Jianling Wang, James Caverlee},
journal={arXiv preprint arXiv:2409.19169},
year={2024},
archivePrefix={arXiv},
eprint={2409.19169},
primaryClass={cs.IR}
}
|
liu2024twincl:
|
arxiv-663011
|
2409.19170
|
An Interactive Hands-Free Controller for a Riding Ballbot to Enable Simple Shared Control Tasks
|
<|reference_start|>An Interactive Hands-Free Controller for a Riding Ballbot to Enable Simple Shared Control Tasks: Our team developed a riding ballbot (called PURE) that is dynamically stable, omnidirectional, and driven by lean-to-steer control. A hands-free admittance control scheme (HACS) was previously integrated to allow riders with different torso functions to control the robot's movements via torso leaning and twisting. Such an interface requires motor coordination skills and could result in collisions with obstacles due to low proficiency. Hence, a shared controller (SC) that limits the speed of PURE could be helpful to ensure the safety of riders. However, the self-balancing dynamics of PURE could result in a weak control authority of its motion, in which the torso motion of the rider could easily result in poor tracking of the command speed dictated by the shared controller. Thus, we proposed an interactive hands-free admittance control scheme (iHACS), which added two modules to HACS to improve the speed-tracking performance of PURE: control gain personalization module and interaction compensation module. Human riding tests of simple tasks, idle-keeping and speed-limiting, were conducted to compare the performance of HACS and iHACS. Two manual wheelchair users and two able-bodied individuals participated in this study. They were instructed to use "adversarial" torso motions that would tax the SC's ability to keep the ballbot idling or below a set speed. In the idle-keeping tasks, iHACS demonstrated minimal translational motion and low command speed tracking RMSE, even with significant torso lean angles. During the speed-limiting task with command speed saturated at 0.5 m/s, the system achieved an average maximum speed of 1.1 m/s with iHACS, compared with that of over 1.9 m/s with HACS. These results suggest that iHACS can enhance PURE's control authority over the rider, which enables PURE to provide physical interactions back to the rider and results in a collaborative rider-robot synergy.<|reference_end|>
|
arxiv
|
@article{xiao2024an,
title={An Interactive Hands-Free Controller for a Riding Ballbot to Enable
Simple Shared Control Tasks},
author={Chenzhang Xiao, Seung Yun Song, Yu Chen, Mahshid Mansouri, Joao Ramos,
William R. Norris, and Elizabeth T. Hsiao-Wecksler},
journal={arXiv preprint arXiv:2409.19170},
year={2024},
archivePrefix={arXiv},
eprint={2409.19170},
primaryClass={cs.RO}
}
|
xiao2024an
|
arxiv-663012
|
2409.19171
|
Reducing Overtreatment of Indeterminate Thyroid Nodules Using a Multimodal Deep Learning Model
|
<|reference_start|>Reducing Overtreatment of Indeterminate Thyroid Nodules Using a Multimodal Deep Learning Model: Objective: Molecular testing (MT) classifies cytologically indeterminate thyroid nodules as benign or malignant with high sensitivity but low positive predictive value (PPV), only using molecular profiles, ignoring ultrasound (US) imaging and biopsy. We address this limitation by applying attention multiple instance learning (AMIL) to US images. Methods: We retrospectively reviewed 333 patients with indeterminate thyroid nodules at UCLA medical center (259 benign, 74 malignant). A multi-modal deep learning AMIL model was developed, combining US images and MT to classify the nodules as benign or malignant and enhance the malignancy risk stratification of MT. Results: The final AMIL model matched MT sensitivity (0.946) while significantly improving PPV (0.477 vs 0.448 for MT alone), indicating fewer false positives while maintaining high sensitivity. Conclusion: Our approach reduces false positives compared to MT while maintaining the same ability to identify positive cases, potentially reducing unnecessary benign thyroid resections in patients with indeterminate nodules.<|reference_end|>
|
arxiv
|
@article{athreya2024reducing,
title={Reducing Overtreatment of Indeterminate Thyroid Nodules Using a
Multimodal Deep Learning Model},
author={Shreeram Athreya, Andrew Melehy, Sujit Silas Armstrong Suthahar,
Vedrana Ivezi'c, Ashwath Radhachandran, Vivek Sant, Chace Moleta, Henry
Zheng, Maitraya Patel, Rinat Masamed, Corey W. Arnold, William Speier},
journal={arXiv preprint arXiv:2409.19171},
year={2024},
archivePrefix={arXiv},
eprint={2409.19171},
primaryClass={q-bio.QM cs.LG eess.IV}
}
|
athreya2024reducing
|
arxiv-663013
|
2409.19172
|
Completely Reachable Almost Group Automata
|
<|reference_start|>Completely Reachable Almost Group Automata: We consider finite deterministic automata such that their alphabets consist of exactly one letter of defect 1 and a set of permutations of the state set. We study under which conditions such an automaton is completely reachable. We focus our attention on the case when the set of permutations generates a transitive imprimitive group.<|reference_end|>
|
arxiv
|
@article{torres2024completely,
title={Completely Reachable Almost Group Automata},
author={David Fernando Casas Torres},
journal={arXiv preprint arXiv:2409.19172},
year={2024},
archivePrefix={arXiv},
eprint={2409.19172},
primaryClass={cs.FL}
}
|
torres2024completely
|
arxiv-663014
|
2409.19173
|
HM3: Heterogeneous Multi-Class Model Merging
|
<|reference_start|>HM3: Heterogeneous Multi-Class Model Merging: Foundation language model deployments often include auxiliary guard-rail models to filter or classify text, detecting jailbreak attempts, biased or toxic output, or ensuring topic adherence. These additional models increase the complexity and cost of model inference, especially since many are also large language models. To address this issue, we explore training-free model merging techniques to consolidate these models into a single, multi-functional model. We propose Heterogeneous Multi-Class Model Merging (HM3) as a simple technique for merging multi-class classifiers with heterogeneous label spaces. Unlike parameter-efficient fine-tuning techniques like LoRA, which require extensive training and add complexity during inference, recent advancements allow models to be merged in a training-free manner. We report promising results for merging BERT-based guard models, some of which attain an average F1-score higher than the source models while reducing the inference time by up to 44%. We introduce self-merging to assess the impact of reduced task-vector density, finding that the more poorly performing hate speech classifier benefits from self-merging while higher-performing classifiers do not, which raises questions about using task vector reduction for model tuning.<|reference_end|>
|
arxiv
|
@article{hackmann2024hm3:,
title={HM3: Heterogeneous Multi-Class Model Merging},
author={Stefan Hackmann},
journal={arXiv preprint arXiv:2409.19173},
year={2024},
archivePrefix={arXiv},
eprint={2409.19173},
primaryClass={cs.CL cs.AI}
}
|
hackmann2024hm3:
|
arxiv-663015
|
2409.19174
|
Feature Estimation of Global Language Processing in EEG Using Attention Maps
|
<|reference_start|>Feature Estimation of Global Language Processing in EEG Using Attention Maps: Understanding the correlation between EEG features and cognitive tasks is crucial for elucidating brain function. Brain activity synchronizes during speaking and listening tasks. However, it is challenging to estimate task-dependent brain activity characteristics with methods with low spatial resolution but high temporal resolution, such as EEG, rather than methods with high spatial resolution, like fMRI. This study introduces a novel approach to EEG feature estimation that utilizes the weights of deep learning models to explore this association. We demonstrate that attention maps generated from Vision Transformers and EEGNet effectively identify features that align with findings from prior studies. EEGNet emerged as the most accurate model regarding subject independence and the classification of Listening and Speaking tasks. The application of Mel-Spectrogram with ViTs enhances the resolution of temporal and frequency-related EEG characteristics. Our findings reveal that the characteristics discerned through attention maps vary significantly based on the input data, allowing for tailored feature extraction from EEG signals. By estimating features, our study reinforces known attributes and predicts new ones, potentially offering fresh perspectives in utilizing EEG for medical purposes, such as early disease detection. These techniques will make substantial contributions to cognitive neuroscience.<|reference_end|>
|
arxiv
|
@article{shimizu2024feature,
title={Feature Estimation of Global Language Processing in EEG Using Attention
Maps},
author={Dai Shimizu, Ko Watanabe, Andreas Dengel},
journal={arXiv preprint arXiv:2409.19174},
year={2024},
archivePrefix={arXiv},
eprint={2409.19174},
primaryClass={q-bio.NC cs.CV eess.SP}
}
|
shimizu2024feature
|
arxiv-663016
|
2409.19176
|
Polynomial Universes and Dependent Types
|
<|reference_start|>Polynomial Universes and Dependent Types: Awodey, later with Newstead, showed how polynomial pseudomonads $(u,1,\Sigma)$ with extra structure (termed "natural models" by Awodey) hold within them the categorical semantics for dependent type theory. Their work presented these ideas clearly but ultimately led them outside of the category of polynomial functors in order to explain all of the structure possessed by such models of type theory. This paper builds off that work -- explicating the categorical semantics of dependent type theory by axiomatizing them \emph{entirely} in the language of polynomial functors. In order to handle the higher-categorical coherences required for such an explanation, we work with polynomial functors internally in the language of Homotopy Type Theory, which allows for higher-dimensional structures such as pseudomonads, etc. to be expressed purely in terms of the structure of a suitably-chosen $\infty$-category of polynomial functors. The move from set theory to Homotopy Type Theory thus has a twofold effect of enabling a simpler exposition of natural models, which is at the same time amenable to formalization in a proof assistant, such as Agda. Moreover, the choice to remain firmly within the setting of polynomial functors reveals many additional structures of natural models that were otherwise left implicit or not considered by Awodey \& Newstead. Chief among these, we highlight the fact that every polynomial pseudomonad $(u,1,\Sigma)$ as above that is also equipped with structure to interpret dependent product types gives rise to a self-distributive law $u \triangleleft u\to u \triangleleft u$, which witnesses the usual distributive law of dependent products over dependent sums.<|reference_end|>
|
arxiv
|
@article{aberlé2024polynomial,
title={Polynomial Universes and Dependent Types},
author={C.B. Aberl'e, David I. Spivak},
journal={arXiv preprint arXiv:2409.19176},
year={2024},
archivePrefix={arXiv},
eprint={2409.19176},
primaryClass={cs.LO cs.PL math.CT}
}
|
aberlé2024polynomial
|
arxiv-663017
|
2409.19177
|
Evidence Is All You Need: Ordering Imaging Studies via Language Model Alignment with the ACR Appropriateness Criteria
|
<|reference_start|>Evidence Is All You Need: Ordering Imaging Studies via Language Model Alignment with the ACR Appropriateness Criteria: Diagnostic imaging studies are an increasingly important component of the workup and management of acutely presenting patients. However, ordering appropriate imaging studies according to evidence-based medical guidelines is a challenging task with a high degree of variability between healthcare providers. To address this issue, recent work has investigated if generative AI and large language models can be leveraged to help clinicians order relevant imaging studies for patients. However, it is challenging to ensure that these tools are correctly aligned with medical guidelines, such as the American College of Radiology's Appropriateness Criteria (ACR AC). In this study, we introduce a framework to intelligently leverage language models by recommending imaging studies for patient cases that are aligned with evidence-based guidelines. We make available a novel dataset of patient "one-liner" scenarios to power our experiments, and optimize state-of-the-art language models to achieve an accuracy on par with clinicians in image ordering. Finally, we demonstrate that our language model-based pipeline can be used as intelligent assistants by clinicians to support image ordering workflows and improve the accuracy of imaging study ordering according to the ACR AC. Our work demonstrates and validates a strategy to leverage AI-based software to improve trustworthy clinical decision making in alignment with expert evidence-based guidelines.<|reference_end|>
|
arxiv
|
@article{yao2024evidence,
title={Evidence Is All You Need: Ordering Imaging Studies via Language Model
Alignment with the ACR Appropriateness Criteria},
author={Michael S. Yao, Allison Chae, Charles E. Kahn Jr., Walter R. Witschey,
James C. Gee, Hersh Sagreiya, Osbert Bastani},
journal={arXiv preprint arXiv:2409.19177},
year={2024},
archivePrefix={arXiv},
eprint={2409.19177},
primaryClass={cs.LG cs.CL cs.CY}
}
|
yao2024evidence
|
arxiv-663018
|
2409.19178
|
FLINT: Learning-based Flow Estimation and Temporal Interpolation for Scientific Ensemble Visualization
|
<|reference_start|>FLINT: Learning-based Flow Estimation and Temporal Interpolation for Scientific Ensemble Visualization: We present FLINT (learning-based FLow estimation and temporal INTerpolation), a novel deep learning-based approach to estimate flow fields for 2D+time and 3D+time scientific ensemble data. FLINT can flexibly handle different types of scenarios with (1) a flow field being partially available for some members (e.g., omitted due to space constraints) or (2) no flow field being available at all (e.g., because it could not be acquired during an experiment). The design of our architecture allows to flexibly cater to both cases simply by adapting our modular loss functions, effectively treating the different scenarios as flow-supervised and flow-unsupervised problems, respectively (with respect to the presence or absence of ground-truth flow). To the best of our knowledge, FLINT is the first approach to perform flow estimation from scientific ensembles, generating a corresponding flow field for each discrete timestep, even in the absence of original flow information. Additionally, FLINT produces high-quality temporal interpolants between scalar fields. FLINT employs several neural blocks, each featuring several convolutional and deconvolutional layers. We demonstrate performance and accuracy for different usage scenarios with scientific ensembles from both simulations and experiments.<|reference_end|>
|
arxiv
|
@article{gadirov2024flint:,
title={FLINT: Learning-based Flow Estimation and Temporal Interpolation for
Scientific Ensemble Visualization},
author={Hamid Gadirov, Jos B.T.M. Roerdink, and Steffen Frey},
journal={arXiv preprint arXiv:2409.19178},
year={2024},
archivePrefix={arXiv},
eprint={2409.19178},
primaryClass={cs.CV}
}
|
gadirov2024flint:
|
arxiv-663019
|
2409.19179
|
A comprehensive review and new taxonomy on superpixel segmentation
|
<|reference_start|>A comprehensive review and new taxonomy on superpixel segmentation: Superpixel segmentation consists of partitioning images into regions composed of similar and connected pixels. Its methods have been widely used in many computer vision applications since it allows for reducing the workload, removing redundant information, and preserving regions with meaningful features. Due to the rapid progress in this area, the literature fails to catch up on more recent works among the compared ones and to categorize the methods according to all existing strategies. This work fills this gap by presenting a comprehensive review with new taxonomy for superpixel segmentation, in which methods are classified according to their processing steps and processing levels of image features. We revisit the recent and popular literature according to our taxonomy and evaluate 20 strategies based on nine criteria: connectivity, compactness, delineation, control over the number of superpixels, color homogeneity, robustness, running time, stability, and visual quality. Our experiments show the trends of each approach in pixel clustering and discuss individual trade-offs. Finally, we provide a new benchmark for superpixel assessment, available at https://github.com/IMScience-PPGINF-PucMinas/superpixel-benchmark.<|reference_end|>
|
arxiv
|
@article{barcelos2024a,
title={A comprehensive review and new taxonomy on superpixel segmentation},
author={I. B. Barcelos, F. de C. Bel'em, L. de M. Jo~ao, Z. K. G. do
Patroc'inio Jr., A. X. Falc~ao, S. J. F. Guimar~aes},
journal={ACM Comput. Surv. 56, 8, Article 200 (August 2024), 39 pages},
year={2024},
doi={10.1145/3652509},
archivePrefix={arXiv},
eprint={2409.19179},
primaryClass={cs.CV}
}
|
barcelos2024a
|
arxiv-663020
|
2409.19180
|
Esports Training, Periodization, and Software -- a Scoping Review
|
<|reference_start|>Esports Training, Periodization, and Software -- a Scoping Review: Electronic sports (esports) and research on this emerging field are interdisciplinary in nature. By extension, it is essential to understand how to standardize and structure training with the help of existing tools developed by years of research in sports sciences and informatics. Our goal in this article was to verify if the current body of research contains substantial evidence of the training systems applied to training esports players. To verify the existing sources, we have applied a framework of scoping review to address the search from multiple scientific databases with further local processing. We conclude that the current research on esports dealt mainly with describing and modeling performance metrics spanned over multiple fragmented research areas (psychology, nutrition, informatics), and yet these building blocks were not assembled into an existing well-functioning theory of performance in esports by providing exercise regimes, and ways of periodization for esports.<|reference_end|>
|
arxiv
|
@article{białecki2024esports,
title={Esports Training, Periodization, and Software -- a Scoping Review},
author={Andrzej Bia{l}ecki and Bart{l}omiej Michalak and Jan Gajewski},
journal={arXiv preprint arXiv:2409.19180},
year={2024},
archivePrefix={arXiv},
eprint={2409.19180},
primaryClass={cs.HC}
}
|
białecki2024esports
|
arxiv-663021
|
2409.19182
|
Artificial-Intelligence Generated Code Considered Harmful: A Road Map for Secure and High-Quality Code Generation
|
<|reference_start|>Artificial-Intelligence Generated Code Considered Harmful: A Road Map for Secure and High-Quality Code Generation: Generating code via a LLM (rather than writing code from scratch), has exploded in popularity. However, the security implications of LLM-generated code are still unknown. We performed a study that compared the security and quality of human-written code with that of LLM-generated code, for a wide range of programming tasks, including data structures, algorithms, cryptographic routines, and LeetCode questions. To assess code security we used unit testing, fuzzing, and static analysis. For code quality, we focused on complexity and size. We found that LLM can generate incorrect code that fails to implement the required functionality, especially for more complicated tasks; such errors can be subtle. For example, for the cryptographic algorithm SHA1, LLM generated an incorrect implementation that nevertheless compiles. In cases where its functionality was correct, we found that LLM-generated code is less secure, primarily due to the lack of defensive programming constructs, which invites a host of security issues such as buffer overflows or integer overflows. Fuzzing has revealed that LLM-generated code is more prone to hangs and crashes than human-written code. Quality-wise, we found that LLM generates bare-bones code that lacks defensive programming constructs, and is typically more complex (per line of code) compared to human-written code. Next, we constructed a feedback loop that asked the LLM to re-generate the code and eliminate the found issues (e.g., malloc overflow, array index out of bounds, null dereferences). We found that the LLM fails to eliminate such issues consistently: while succeeding in some cases, we found instances where the re-generated, supposedly more secure code, contains new issues; we also found that upon prompting, LLM can introduce issues in files that were issues-free before prompting.<|reference_end|>
|
arxiv
|
@article{chong2024artificial-intelligence,
title={Artificial-Intelligence Generated Code Considered Harmful: A Road Map
for Secure and High-Quality Code Generation},
author={Chun Jie Chong, Zhihao Yao, Iulian Neamtiu},
journal={arXiv preprint arXiv:2409.19182},
year={2024},
archivePrefix={arXiv},
eprint={2409.19182},
primaryClass={cs.CR cs.AI}
}
|
chong2024artificial-intelligence
|
arxiv-663022
|
2409.19184
|
Learning-Based Image Compression for Machines
|
<|reference_start|>Learning-Based Image Compression for Machines: While learning based compression techniques for images have outperformed traditional methods, they have not been widely adopted in machine learning pipelines. This is largely due to lack of standardization and lack of retention of salient features needed for such tasks. Decompression of images have taken a back seat in recent years while the focus has shifted to an image's utility in performing machine learning based analysis on top of them. Thus the demand for compression pipelines that incorporate such features from images has become ever present. The methods outlined in the report build on the recent work done on learning based image compression techniques to incorporate downstream tasks in them. We propose various methods of finetuning and enhancing different parts of pretrained compression encoding pipeline and present the results of our investigation regarding the performance of vision tasks using compression based pipelines.<|reference_end|>
|
arxiv
|
@article{gupta2024learning-based,
title={Learning-Based Image Compression for Machines},
author={Kartik Gupta, Kimberley Faria, Vikas Mehta},
journal={arXiv preprint arXiv:2409.19184},
year={2024},
archivePrefix={arXiv},
eprint={2409.19184},
primaryClass={eess.IV cs.CV cs.LG}
}
|
gupta2024learning-based
|
arxiv-663023
|
2409.19185
|
Semi-Supervised Bone Marrow Lesion Detection from Knee MRI Segmentation Using Mask Inpainting Models
|
<|reference_start|>Semi-Supervised Bone Marrow Lesion Detection from Knee MRI Segmentation Using Mask Inpainting Models: Bone marrow lesions (BMLs) are critical indicators of knee osteoarthritis (OA). Since they often appear as small, irregular structures with indistinguishable edges in knee magnetic resonance images (MRIs), effective detection of BMLs in MRI is vital for OA diagnosis and treatment. This paper proposes a semi-supervised local anomaly detection method using mask inpainting models for identification of BMLs in high-resolution knee MRI, effectively integrating a 3D femur bone segmentation model, a large mask inpainting model, and a series of post-processing techniques. The method was evaluated using MRIs at various resolutions from a subset of the public Osteoarthritis Initiative database. Dice score, Intersection over Union (IoU), and pixel-level sensitivity, specificity, and accuracy showed an advantage over the multiresolution knowledge distillation method-a state-of-the-art global anomaly detection method. Especially, segmentation performance is enhanced on higher-resolution images, achieving an over two times performance increase on the Dice score and the IoU score at a 448x448 resolution level. We also demonstrate that with increasing size of the BML region, both the Dice and IoU scores improve as the proportion of distinguishable boundary decreases. The identified BML masks can serve as markers for downstream tasks such as segmentation and classification. The proposed method has shown a potential in improving BML detection, laying a foundation for further advances in imaging-based OA research.<|reference_end|>
|
arxiv
|
@article{qin2024semi-supervised,
title={Semi-Supervised Bone Marrow Lesion Detection from Knee MRI Segmentation
Using Mask Inpainting Models},
author={Shihua Qin, Ming Zhang, Juan Shan, Taehoon Shin, Jonghye Woo, Fangxu
Xing},
journal={arXiv preprint arXiv:2409.19185},
year={2024},
archivePrefix={arXiv},
eprint={2409.19185},
primaryClass={eess.IV cs.AI cs.CV}
}
|
qin2024semi-supervised
|
arxiv-663024
|
2409.19189
|
State estimation for parallel-connected batteries via inverse dynamic modeling
|
<|reference_start|>State estimation for parallel-connected batteries via inverse dynamic modeling: This paper examines the problem of estimating the states, including state of charge, of battery cells connected in parallel. Previous research highlights the importance of this problem, and presents multiple approaches for solving it. Algorithm scalability and observability analysis can both be challenging, particularly because the underlying pack dynamics are governed by differential algebraic equations. Our work addresses these challenges from a novel perspective that begins by inverting the causality of parallel pack dynamics, which breaks the pack model's underlying algebraic loop. This simplifies observability analysis and observer design significantly, leading to three novel contributions. First, the paper derives mathematical conditions for state observability that apply regardless of the number of battery cells and the order of their individual dynamics. Second, the paper presents an approach for grouping battery cells such that their lumped dynamics are observable. Finally, the paper presents a novel pack state estimator that achieves computational tractability by employing inverse dynamic modeling. We conclude by presenting a Monte Carlo simulation study of this estimator using experimentally-parameterized models of two battery chemistries. The simulation results highlight the computational benefits of both the clustering strategy and inverse dynamics approach for state estimation.<|reference_end|>
|
arxiv
|
@article{lee2024state,
title={State estimation for parallel-connected batteries via inverse dynamic
modeling},
author={Hannah Lee, Casey Casten, Hosam Fathy},
journal={arXiv preprint arXiv:2409.19189},
year={2024},
archivePrefix={arXiv},
eprint={2409.19189},
primaryClass={eess.SY cs.SY}
}
|
lee2024state
|
arxiv-663025
|
2409.19190
|
RAIL: Reachability-Aided Imitation Learning for Safe Policy Execution
|
<|reference_start|>RAIL: Reachability-Aided Imitation Learning for Safe Policy Execution: Imitation learning (IL) has shown great success in learning complex robot manipulation tasks. However, there remains a need for practical safety methods to justify widespread deployment. In particular, it is important to certify that a system obeys hard constraints on unsafe behavior in settings when it is unacceptable to design a tradeoff between performance and safety via tuning the policy (i.e. soft constraints). This leads to the question, how does enforcing hard constraints impact the performance (meaning safely completing tasks) of an IL policy? To answer this question, this paper builds a reachability-based safety filter to enforce hard constraints on IL, which we call Reachability-Aided Imitation Learning (RAIL). Through evaluations with state-of-the-art IL policies in mobile robots and manipulation tasks, we make two key findings. First, the highest-performing policies are sometimes only so because they frequently violate constraints, and significantly lose performance under hard constraints. Second, surprisingly, hard constraints on the lower-performing policies can occasionally increase their ability to perform tasks safely. Finally, hardware evaluation confirms the method can operate in real time.<|reference_end|>
|
arxiv
|
@article{jung2024rail:,
title={RAIL: Reachability-Aided Imitation Learning for Safe Policy Execution},
author={Wonsuhk Jung, Dennis Anthony, Utkarsh A. Mishra, Nadun Ranawaka
Arachchige, Matthew Bronars, Danfei Xu, Shreyas Kousik},
journal={arXiv preprint arXiv:2409.19190},
year={2024},
archivePrefix={arXiv},
eprint={2409.19190},
primaryClass={cs.RO}
}
|
jung2024rail:
|
arxiv-663026
|
2409.19192
|
An Extension of the Euler-Maclaurin Summation Formula to Nearly Singular Functions
|
<|reference_start|>An Extension of the Euler-Maclaurin Summation Formula to Nearly Singular Functions: A extension of the Euler-Maclaurin (E-M) formula to near-singular functions is presented. This extension is derived based on earlier generalized E-M formulas for singular functions. The new E-M formulas consists of two components: a ``singular'' component that is a continuous extension of the earlier singular E-M formulas, and a ``jump'' component associated with the discontinuity of the integral with respect to a parameter that controls near singularity. The singular component of the new E-M formulas is an asymptotic series whose coefficients depend on the Hurwitz zeta function or the digamma function. Numerical examples of near-singular quadrature based on the extended E-M formula are presented, where accuracies of machine precision are achieved insensitive to the strength of the near singularity and with a very small number of quadrature nodes.<|reference_end|>
|
arxiv
|
@article{wu2024an,
title={An Extension of the Euler-Maclaurin Summation Formula to Nearly Singular
Functions},
author={Bowei Wu},
journal={arXiv preprint arXiv:2409.19192},
year={2024},
archivePrefix={arXiv},
eprint={2409.19192},
primaryClass={math.NA cs.NA}
}
|
wu2024an
|
arxiv-663027
|
2409.19194
|
An In-depth Analysis of a Nation-Sponsored Attack: Case Study and Cybersecurity Insights
|
<|reference_start|>An In-depth Analysis of a Nation-Sponsored Attack: Case Study and Cybersecurity Insights: Nation-sponsored cyberattacks pose a significant threat to national security by targeting critical infrastructure and disrupting essential services. One of the most impactful cyber threats affecting South Korea's banking sector and infrastructure was the Dark Seoul cyberattack, which occurred several years ago. Believed to have been orchestrated by North Korean state-sponsored hackers, the attack employed spear phishing, DNS poisoning, and malware to compromise systems, causing widespread disruption. In this paper, we conduct an in-depth analysis of the Dark Seoul attack, examining the techniques used and providing insights and defense recommendations for the global cybersecurity community. The motivations behind the attack are explored, along with an assessment of South Korea's response and the broader implications for cybersecurity policy. Our analysis highlights the vulnerabilities exploited and underscores the need for more proactive defenses against state-sponsored cyber threats. This paper, therefore, emphasizes the critical need for stronger national cybersecurity defenses in the face of such threats.<|reference_end|>
|
arxiv
|
@article{pakshad2024an,
title={An In-depth Analysis of a Nation-Sponsored Attack: Case Study and
Cybersecurity Insights},
author={Puya Pakshad, Abiha Hussain, Maks Dudek, Leen Mobarki, Abel Castilla},
journal={arXiv preprint arXiv:2409.19194},
year={2024},
archivePrefix={arXiv},
eprint={2409.19194},
primaryClass={cs.CR cs.SE}
}
|
pakshad2024an
|
arxiv-663028
|
2409.19200
|
Faster Acceleration for Steepest Descent
|
<|reference_start|>Faster Acceleration for Steepest Descent: We propose a new accelerated first-order method for convex optimization under non-Euclidean smoothness assumptions. In contrast to standard acceleration techniques, our approach uses primal-dual iterate sequences taken with respect to differing norms, which are then coupled using an implicitly determined interpolation parameter. For $\ell_p$ norm smooth problems in $d$ dimensions, our method provides an iteration complexity improvement of up to $O(d^{1-\frac{2}{p}})$ in terms of calls to a first-order oracle, thereby allowing us to circumvent long-standing barriers in accelerated non-Euclidean steepest descent.<|reference_end|>
|
arxiv
|
@article{bai2024faster,
title={Faster Acceleration for Steepest Descent},
author={Site Bai and Brian Bullins},
journal={arXiv preprint arXiv:2409.19200},
year={2024},
archivePrefix={arXiv},
eprint={2409.19200},
primaryClass={math.OC cs.LG stat.ML}
}
|
bai2024faster
|
arxiv-663029
|
2409.19202
|
Safe Delay-Adaptive Control of Strict-Feedback Nonlinear Systems with Application in Vehicle Platooning
|
<|reference_start|>Safe Delay-Adaptive Control of Strict-Feedback Nonlinear Systems with Application in Vehicle Platooning: This paper presents a safe delay-adaptive control for a strict-feedback nonlinear ODE with a delayed actuator, whose dynamic is also a strict-feedback nonlinear ODE and the delay length is unknown. By formulating the delay as a transport PDE, the plant becomes a sandwich configuration consisting of nonlinear ODE-transport PDE-nonlinear ODE, where the transport speed in the PDE is unknown. We propose a predictor-based nonovershooting backstepping transformation to build the nominal safe delay-compensated control, guaranteeing that the output of the distal ODE safely tracks the target trajectory from one side without undershooting. To address the uncertainty in the delay, we incorporate recent delay-adaptive and safe adaptive technologies to build a safe adaptive-delay controller. The adaptive closed-loop system ensures 1) the exact identification of the unknown delay in finite time; 2) the output state stays in the safe region all the time, especially in the original safe region, instead of a subset, after a finite time; 3) all states are bounded, and moreover, they will converge to zero if the target trajectory is identically zero. In the simulation, the proposed control design is verified in the application of safe vehicle platooning. It regulates the spacing between adjacent vehicles to converge to a small distance and avoids collisions by ensuring they do not breach the safe distance at any time, even in the presence of large unknown delays and at a relatively high speed.<|reference_end|>
|
arxiv
|
@article{zhao2024safe,
title={Safe Delay-Adaptive Control of Strict-Feedback Nonlinear Systems with
Application in Vehicle Platooning},
author={Zhenxu Zhao and Ji Wang},
journal={arXiv preprint arXiv:2409.19202},
year={2024},
archivePrefix={arXiv},
eprint={2409.19202},
primaryClass={eess.SY cs.SY}
}
|
zhao2024safe
|
arxiv-663030
|
2409.19209
|
Boosting SISSO Performance on Small Sample Datasets by Using Random Forests Prescreening for Complex Feature Selection
|
<|reference_start|>Boosting SISSO Performance on Small Sample Datasets by Using Random Forests Prescreening for Complex Feature Selection: In materials science, data-driven methods accelerate material discovery and optimization while reducing costs and improving success rates. Symbolic regression is a key to extracting material descriptors from large datasets, in particular the Sure Independence Screening and Sparsifying Operator (SISSO) method. While SISSO needs to store the entire expression space to impose heavy memory demands, it limits the performance in complex problems. To address this issue, we propose a RF-SISSO algorithm by combining Random Forests (RF) with SISSO. In this algorithm, the Random Forest algorithm is used for prescreening, capturing non-linear relationships and improving feature selection, which may enhance the quality of the input data and boost the accuracy and efficiency on regression and classification tasks. For a testing on the SISSO's verification problem for 299 materials, RF-SISSO demonstrates its robust performance and high accuracy. RF-SISSO can maintain the testing accuracy above 0.9 across all four training sample sizes and significantly enhancing regression efficiency, especially in training subsets with smaller sample sizes. For the training subset with 45 samples, the efficiency of RF-SISSO was 265 times higher than that of original SISSO. As collecting large datasets would be both costly and time-consuming in the practical experiments, it is thus believed that RF-SISSO may benefit scientific researches by offering a high predicting accuracy with limited data efficiently.<|reference_end|>
|
arxiv
|
@article{jiang2024boosting,
title={Boosting SISSO Performance on Small Sample Datasets by Using Random
Forests Prescreening for Complex Feature Selection},
author={Xiaolin Jiang, Guanqi Liu, Jiaying Xie, Zhenpeng Hu},
journal={arXiv preprint arXiv:2409.19209},
year={2024},
archivePrefix={arXiv},
eprint={2409.19209},
primaryClass={cs.LG cond-mat.mtrl-sci physics.data-an}
}
|
jiang2024boosting
|
arxiv-663031
|
2409.19210
|
Learning to Obstruct Few-Shot Image Classification over Restricted Classes
|
<|reference_start|>Learning to Obstruct Few-Shot Image Classification over Restricted Classes: Advancements in open-source pre-trained backbones make it relatively easy to fine-tune a model for new tasks. However, this lowered entry barrier poses potential risks, e.g., bad actors developing models for harmful applications. A question arises: Is possible to develop a pre-trained model that is difficult to fine-tune for certain downstream tasks? To begin studying this, we focus on few-shot classification (FSC). Specifically, we investigate methods to make FSC more challenging for a set of restricted classes while maintaining the performance of other classes. We propose to meta-learn over the pre-trained backbone in a manner that renders it a ''poor initialization''. Our proposed Learning to Obstruct (LTO) algorithm successfully obstructs four FSC methods across three datasets, including ImageNet and CIFAR100 for image classification, as well as CelebA for attribute classification.<|reference_end|>
|
arxiv
|
@article{zheng2024learning,
title={Learning to Obstruct Few-Shot Image Classification over Restricted
Classes},
author={Amber Yijia Zheng, Chiao-An Yang and Raymond A. Yeh},
journal={arXiv preprint arXiv:2409.19210},
year={2024},
archivePrefix={arXiv},
eprint={2409.19210},
primaryClass={cs.CV}
}
|
zheng2024learning
|
arxiv-663032
|
2409.19211
|
Programming with High-Level Abstractions, Proceedings of the 3rd Workshop on Logic and Practice of Programming
|
<|reference_start|>Programming with High-Level Abstractions, Proceedings of the 3rd Workshop on Logic and Practice of Programming: This proceedings contains abstracts and position papers for the work presented at the third Logic and Practice of Programming (LPOP) Workshop. The workshop was held online, using zoom, at stonybrook.zoom.us, on December 13, 2022. The workshop focused on core high-level abstractions around sets and logic rules, to help bring them to the general practice of programming.<|reference_end|>
|
arxiv
|
@article{warren2024programming,
title={Programming with High-Level Abstractions, Proceedings of the 3rd
Workshop on Logic and Practice of Programming},
author={David S. Warren and Yanhong A. Liu},
journal={arXiv preprint arXiv:2409.19211},
year={2024},
archivePrefix={arXiv},
eprint={2409.19211},
primaryClass={cs.PL}
}
|
warren2024programming
|
arxiv-663033
|
2409.19212
|
An Accelerated Algorithm for Stochastic Bilevel Optimization under Unbounded Smoothness
|
<|reference_start|>An Accelerated Algorithm for Stochastic Bilevel Optimization under Unbounded Smoothness: This paper investigates a class of stochastic bilevel optimization problems where the upper-level function is nonconvex with potentially unbounded smoothness and the lower-level problem is strongly convex. These problems have significant applications in sequential data learning, such as text classification using recurrent neural networks. The unbounded smoothness is characterized by the smoothness constant of the upper-level function scaling linearly with the gradient norm, lacking a uniform upper bound. Existing state-of-the-art algorithms require $\widetilde{O}(1/\epsilon^4)$ oracle calls of stochastic gradient or Hessian/Jacobian-vector product to find an $\epsilon$-stationary point. However, it remains unclear if we can further improve the convergence rate when the assumptions for the function in the population level also hold for each random realization almost surely (e.g., Lipschitzness of each realization of the stochastic gradient). To address this issue, we propose a new Accelerated Bilevel Optimization algorithm named AccBO. The algorithm updates the upper-level variable by normalized stochastic gradient descent with recursive momentum and the lower-level variable by the stochastic Nesterov accelerated gradient descent algorithm with averaging. We prove that our algorithm achieves an oracle complexity of $\widetilde{O}(1/\epsilon^3)$ to find an $\epsilon$-stationary point. Our proof relies on a novel lemma characterizing the dynamics of stochastic Nesterov accelerated gradient descent algorithm under distribution drift with high probability for the lower-level variable, which is of independent interest and also plays a crucial role in analyzing the hypergradient estimation error over time. Experimental results on various tasks confirm that our proposed algorithm achieves the predicted theoretical acceleration and significantly outperforms baselines in bilevel optimization.<|reference_end|>
|
arxiv
|
@article{gong2024an,
title={An Accelerated Algorithm for Stochastic Bilevel Optimization under
Unbounded Smoothness},
author={Xiaochuan Gong, Jie Hao, Mingrui Liu},
journal={arXiv preprint arXiv:2409.19212},
year={2024},
archivePrefix={arXiv},
eprint={2409.19212},
primaryClass={cs.LG math.OC}
}
|
gong2024an
|
arxiv-663034
|
2409.19213
|
Feature-Prescribed Iterative Learning Control of Waggle Dance Movement for Social Motor Coordination in Joint Actions
|
<|reference_start|>Feature-Prescribed Iterative Learning Control of Waggle Dance Movement for Social Motor Coordination in Joint Actions: Extensive experiments suggest that motor coordination among human participants may contribute to social affinity and emotional attachment, which has great potential in the clinical treatment of social disorders or schizophrenia. Mirror game provides an effective experimental paradigm for studying social motor coordination. Nevertheless, the lack of movement richness prevents the emergence of high-level coordination in the existing one-dimensional experiments. To tackle this problem, this work develops a two-dimensional experimental paradigm of mirror game by playing waggle dance between two participants. In particular, an online control architecture of customized virtual player is created to coordinate with human player. Therein, an iterative learning control algorithm is proposed by integrating position tracking and behavior imitation with prescribed kinematic feature. Moreover, convergence analysis of control algorithm is conducted to guarantee the online performance of virtual player. Finally, the proposed control strategy is validated by matching experimental data and compared with other control methods using a set of performance indexes.<|reference_end|>
|
arxiv
|
@article{guo2024feature-prescribed,
title={Feature-Prescribed Iterative Learning Control of Waggle Dance Movement
for Social Motor Coordination in Joint Actions},
author={Bowen Guo, Chao Zhai},
journal={arXiv preprint arXiv:2409.19213},
year={2024},
archivePrefix={arXiv},
eprint={2409.19213},
primaryClass={cs.HC}
}
|
guo2024feature-prescribed
|
arxiv-663035
|
2409.19214
|
Group Distributionally Robust Optimization can Suppress Class Imbalance Effect in Network Traffic Classification
|
<|reference_start|>Group Distributionally Robust Optimization can Suppress Class Imbalance Effect in Network Traffic Classification: Internet services have led to the eruption of traffic, and machine learning on these Internet data has become an indispensable tool, especially when the application is risk-sensitive. This paper focuses on network traffic classification in the presence of class imbalance, which fundamentally and ubiquitously exists in Internet data analysis. This existence of class imbalance mostly drifts the optimal decision boundary, resulting in a less optimal solution for machine learning models. To alleviate the effect, we propose to design strategies for alleviating the class imbalance through the lens of group distributionally robust optimization. Our approach iteratively updates the non-parametric weights for separate classes and optimizes the learning model by minimizing reweighted losses. We interpret the optimization steps from a Stackelberg game and perform extensive experiments on typical benchmarks. Results show that our approach can not only suppress the negative effect of class imbalance but also improve the comprehensive performance in prediction.<|reference_end|>
|
arxiv
|
@article{du2024group,
title={Group Distributionally Robust Optimization can Suppress Class Imbalance
Effect in Network Traffic Classification},
author={Wumei Du, Qi Wang, Yiqin Lv, Dong Liang, Guanlin Wu, Xingxing Liang,
Zheng Xie},
journal={arXiv preprint arXiv:2409.19214},
year={2024},
archivePrefix={arXiv},
eprint={2409.19214},
primaryClass={stat.ML cs.LG}
}
|
du2024group
|
arxiv-663036
|
2409.19215
|
1st Place Solution to the 8th HANDS Workshop Challenge -- ARCTIC Track: 3DGS-based Bimanual Category-agnostic Interaction Reconstruction
|
<|reference_start|>1st Place Solution to the 8th HANDS Workshop Challenge -- ARCTIC Track: 3DGS-based Bimanual Category-agnostic Interaction Reconstruction: This report describes our 1st place solution to the 8th HANDS workshop challenge (ARCTIC track) in conjunction with ECCV 2024. In this challenge, we address the task of bimanual category-agnostic hand-object interaction reconstruction, which aims to generate 3D reconstructions of both hands and the object from a monocular video, without relying on predefined templates. This task is particularly challenging due to the significant occlusion and dynamic contact between the hands and the object during bimanual manipulation. We worked to resolve these issues by introducing a mask loss and a 3D contact loss, respectively. Moreover, we applied 3D Gaussian Splatting (3DGS) to this task. As a result, our method achieved a value of 38.69 in the main metric, CD$_h$, on the ARCTIC test set.<|reference_end|>
|
arxiv
|
@article{on20241st,
title={1st Place Solution to the 8th HANDS Workshop Challenge -- ARCTIC Track:
3DGS-based Bimanual Category-agnostic Interaction Reconstruction},
author={Jeongwan On, Kyeonghwan Gwak, Gunyoung Kang, Hyein Hwang, Soohyun
Hwang, Junuk Cha, Jaewook Han, Seungryul Baek},
journal={arXiv preprint arXiv:2409.19215},
year={2024},
archivePrefix={arXiv},
eprint={2409.19215},
primaryClass={cs.CV}
}
|
on20241st
|
arxiv-663037
|
2409.19218
|
A Characterization of List Regression
|
<|reference_start|>A Characterization of List Regression: There has been a recent interest in understanding and characterizing the sample complexity of list learning tasks, where the learning algorithm is allowed to make a short list of $k$ predictions, and we simply require one of the predictions to be correct. This includes recent works characterizing the PAC sample complexity of standard list classification and online list classification. Adding to this theme, in this work, we provide a complete characterization of list PAC regression. We propose two combinatorial dimensions, namely the $k$-OIG dimension and the $k$-fat-shattering dimension, and show that they optimally characterize realizable and agnostic $k$-list regression respectively. These quantities generalize known dimensions for standard regression. Our work thus extends existing list learning characterizations from classification to regression.<|reference_end|>
|
arxiv
|
@article{pabbaraju2024a,
title={A Characterization of List Regression},
author={Chirag Pabbaraju, Sahasrajit Sarmasarkar},
journal={arXiv preprint arXiv:2409.19218},
year={2024},
archivePrefix={arXiv},
eprint={2409.19218},
primaryClass={cs.LG cs.DS stat.ML}
}
|
pabbaraju2024a
|
arxiv-663038
|
2409.19219
|
Sharing-Based Channel Access Procedure For Next Generation of Wireless LAN
|
<|reference_start|>Sharing-Based Channel Access Procedure For Next Generation of Wireless LAN: This paper proposes a new channel access procedure to mitigate the channel access contention in next generation of Wireless Local-Area Networks (WLANs) by allowing cooperation among devices belonging to same network, while maintaining high flexibility in terms of how each device may contend the medium. After introducing the details of the proposed procedure, which is here referred to as sharing-based protocol, an analytical analysis is provided to compare it with the two state-of-art protocols currently adopted in IEEE 802.11 standard, i.e, Enhanced Distributed Channel Access (EDCA)-based and trigger-based protocol. In this regards, closed form expressions are derived to evaluate the success probability of channel access for each protocol. In order to show the merit of the proposed procedure, a comprehensive system level analysis is also provided, which highlights that the proposed procedure outperforms the two state-of-art protocols in terms of mitigating the End-to-End (E2E) delay and allowing a better spectrum utilization by reducing the overall congestion in the system.<|reference_end|>
|
arxiv
|
@article{xia2024sharing-based,
title={Sharing-Based Channel Access Procedure For Next Generation of Wireless
LAN},
author={Qing Xia and Salvatore Talarico},
journal={arXiv preprint arXiv:2409.19219},
year={2024},
archivePrefix={arXiv},
eprint={2409.19219},
primaryClass={cs.NI cs.ET cs.IT math.IT}
}
|
xia2024sharing-based
|
arxiv-663039
|
2409.19220
|
Extending Depth of Field for Varifocal Multiview Images
|
<|reference_start|>Extending Depth of Field for Varifocal Multiview Images: Optical imaging systems are generally limited by the depth of field because of the nature of the optics. Therefore, extending depth of field (EDoF) is a fundamental task for meeting the requirements of emerging visual applications. To solve this task, the common practice is using multi-focus images from a single viewpoint. This method can obtain acceptable quality of EDoF under the condition of fixed field of view, but it is only applicable to static scenes and the field of view is limited and fixed. An emerging data type, varifocal multiview images have the potential to become a new paradigm for solving the EDoF, because the data contains more field of view information than multi-focus images. To realize EDoF of varifocal multiview images, we propose an end-to-end method for the EDoF, including image alignment, image optimization and image fusion. Experimental results demonstrate the efficiency of the proposed method.<|reference_end|>
|
arxiv
|
@article{li2024extending,
title={Extending Depth of Field for Varifocal Multiview Images},
author={Zhilong Li, Kejun Wu, Qiong Liu, and You Yang},
journal={arXiv preprint arXiv:2409.19220},
year={2024},
archivePrefix={arXiv},
eprint={2409.19220},
primaryClass={cs.CV cs.MM}
}
|
li2024extending
|
arxiv-663040
|
2409.19221
|
Cauchy activation function and XNet
|
<|reference_start|>Cauchy activation function and XNet: We have developed a novel activation function, named the Cauchy Activation Function. This function is derived from the Cauchy Integral Theorem in complex analysis and is specifically tailored for problems requiring high precision. This innovation has led to the creation of a new class of neural networks, which we call (Comple)XNet, or simply XNet. We will demonstrate that XNet is particularly effective for high-dimensional challenges such as image classification and solving Partial Differential Equations (PDEs). Our evaluations show that XNet significantly outperforms established benchmarks like MNIST and CIFAR-10 in computer vision, and offers substantial advantages over Physics-Informed Neural Networks (PINNs) in both low-dimensional and high-dimensional PDE scenarios.<|reference_end|>
|
arxiv
|
@article{li2024cauchy,
title={Cauchy activation function and XNet},
author={Xin Li, Zhihong Xia, Hongkun Zhang},
journal={arXiv preprint arXiv:2409.19221},
year={2024},
archivePrefix={arXiv},
eprint={2409.19221},
primaryClass={cs.LG cs.CV cs.NE}
}
|
li2024cauchy
|
arxiv-663041
|
2409.19222
|
How do Practitioners Perceive Energy Consumption on Stack Overflow?
|
<|reference_start|>How do Practitioners Perceive Energy Consumption on Stack Overflow?: Energy consumption of software applications has emerged as a critical issue for practitioners to contemplate in their daily development processes. Previous studies have performed user surveys with a limited number of practitioners to comprehend practitioners' viewpoints on energy consumption. In this paper, we complement prior studies by conducting an empirical analysis of a meticulously curated dataset comprising 985 Stack Overflow (SO) questions concerning energy consumption. These questions reflect real-world energy-related predicaments faced by practitioners in their daily development activities. To understand practitioners' perception of energy consumption, we investigate the intentions behind these questions, their semantic topics, as well as the tag categories associated with these questions. Our empirical study results reveal that (i) the intentions that drive the questioners to initiate posts and ask questions are primarily associated with understanding a concept or how to use an API; (ii) the most prevalent topic related to energy consumption concerns computing resources; (iii) monitoring energy usage poses a challenging issue, and it takes the longest response time to receive a community response to the questions; and (iv) practitioners are apprehensive about energy consumption from different levels, i.e., hardware, operating systems, and programming languages, during the development of the applications. Our work furnishes insights into the issues related to energy consumption faced by practitioners. Our observations raise awareness among practitioners about the impact of energy consumption on developing software systems from different perspectives, such as coding efficiency and energy monitoring, and shed light on future research opportunities to assist practitioners in developing energy-efficient software systems.<|reference_end|>
|
arxiv
|
@article{jin2024how,
title={How do Practitioners Perceive Energy Consumption on Stack Overflow?},
author={Bihui Jin, Heng Li, Ying Zou},
journal={arXiv preprint arXiv:2409.19222},
year={2024},
archivePrefix={arXiv},
eprint={2409.19222},
primaryClass={cs.SE cs.PF}
}
|
jin2024how
|
arxiv-663042
|
2409.19223
|
Summit Vitals: Multi-Camera and Multi-Signal Biosensing at High Altitudes
|
<|reference_start|>Summit Vitals: Multi-Camera and Multi-Signal Biosensing at High Altitudes: Video photoplethysmography (vPPG) is an emerging method for non-invasive and convenient measurement of physiological signals, utilizing two primary approaches: remote video PPG (rPPG) and contact video PPG (cPPG). Monitoring vitals in high-altitude environments, where heart rates tend to increase and blood oxygen levels often decrease, presents significant challenges. To address these issues, we introduce the SUMS dataset comprising 80 synchronized non-contact facial and contact finger videos from 10 subjects during exercise and oxygen recovery scenarios, capturing PPG, respiration rate (RR), and SpO2. This dataset is designed to validate video vitals estimation algorithms and compare facial rPPG with finger cPPG. Additionally, fusing videos from different positions (i.e., face and finger) reduces the mean absolute error (MAE) of SpO2 predictions by 7.6\% and 10.6\% compared to only face and only finger, respectively. In cross-subject evaluation, we achieve an MAE of less than 0.5 BPM for HR estimation and 2.5\% for SpO2 estimation, demonstrating the precision of our multi-camera fusion techniques. Our findings suggest that simultaneous training on multiple indicators, such as PPG and blood oxygen, can reduce MAE in SpO2 estimation by 17.8\%.<|reference_end|>
|
arxiv
|
@article{liu2024summit,
title={Summit Vitals: Multi-Camera and Multi-Signal Biosensing at High
Altitudes},
author={Ke Liu, Jiankai Tang, Zhang Jiang, Yuntao Wang, Xiaojing Liu, Dong Li,
Yuanchun Shi},
journal={arXiv preprint arXiv:2409.19223},
year={2024},
archivePrefix={arXiv},
eprint={2409.19223},
primaryClass={cs.CV eess.SP}
}
|
liu2024summit
|
arxiv-663043
|
2409.19226
|
Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and Reinforcement Learning
|
<|reference_start|>Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and Reinforcement Learning: The real world is unpredictable. Therefore, to solve long-horizon decision-making problems with autonomous robots, we must construct agents that are capable of adapting to changes in the environment during deployment. Model-based planning approaches can enable robots to solve complex, long-horizon tasks in a variety of environments. However, such approaches tend to be brittle when deployed into an environment featuring a novel situation that their underlying model does not account for. In this work, we propose to learn a ``bridge policy'' via Reinforcement Learning (RL) to adapt to such novelties. We introduce a simple formulation for such learning, where the RL problem is constructed with a special ``CallPlanner'' action that terminates the bridge policy and hands control of the agent back to the planner. This allows the RL policy to learn the set of states in which querying the planner and following the returned plan will achieve the goal. We show that this formulation enables the agent to rapidly learn by leveraging the planner's knowledge to avoid challenging long-horizon exploration caused by sparse reward. In experiments across three different simulated domains of varying complexity, we demonstrate that our approach is able to learn policies that adapt to novelty more efficiently than several baselines, including a pure RL baseline. We also demonstrate that the learned bridge policy is generalizable in that it can be combined with the planner to enable the agent to solve more complex tasks with multiple instances of the encountered novelty.<|reference_end|>
|
arxiv
|
@article{li2024learning,
title={Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and
Reinforcement Learning},
author={Alicia Li, Nishanth Kumar, Tom'as Lozano-P'erez, Leslie Kaelbling},
journal={arXiv preprint arXiv:2409.19226},
year={2024},
archivePrefix={arXiv},
eprint={2409.19226},
primaryClass={cs.RO cs.AI}
}
|
li2024learning
|
arxiv-663044
|
2409.19228
|
GS-EVT: Cross-Modal Event Camera Tracking based on Gaussian Splatting
|
<|reference_start|>GS-EVT: Cross-Modal Event Camera Tracking based on Gaussian Splatting: Reliable self-localization is a foundational skill for many intelligent mobile platforms. This paper explores the use of event cameras for motion tracking thereby providing a solution with inherent robustness under difficult dynamics and illumination. In order to circumvent the challenge of event camera-based mapping, the solution is framed in a cross-modal way. It tracks a map representation that comes directly from frame-based cameras. Specifically, the proposed method operates on top of gaussian splatting, a state-of-the-art representation that permits highly efficient and realistic novel view synthesis. The key of our approach consists of a novel pose parametrization that uses a reference pose plus first order dynamics for local differential image rendering. The latter is then compared against images of integrated events in a staggered coarse-to-fine optimization scheme. As demonstrated by our results, the realistic view rendering ability of gaussian splatting leads to stable and accurate tracking across a variety of both publicly available and newly recorded data sequences.<|reference_end|>
|
arxiv
|
@article{liu2024gs-evt:,
title={GS-EVT: Cross-Modal Event Camera Tracking based on Gaussian Splatting},
author={Tao Liu, Runze Yuan, Yi'ang Ju, Xun Xu, Jiaqi Yang, Xiangting Meng,
Xavier Lagorce, Laurent Kneip},
journal={arXiv preprint arXiv:2409.19228},
year={2024},
archivePrefix={arXiv},
eprint={2409.19228},
primaryClass={cs.CV}
}
|
liu2024gs-evt:
|
arxiv-663045
|
2409.19231
|
Double Actor-Critic with TD Error-Driven Regularization in Reinforcement Learning
|
<|reference_start|>Double Actor-Critic with TD Error-Driven Regularization in Reinforcement Learning: To obtain better value estimation in reinforcement learning, we propose a novel algorithm based on the double actor-critic framework with temporal difference error-driven regularization, abbreviated as TDDR. TDDR employs double actors, with each actor paired with a critic, thereby fully leveraging the advantages of double critics. Additionally, TDDR introduces an innovative critic regularization architecture. Compared to classical deterministic policy gradient-based algorithms that lack a double actor-critic structure, TDDR provides superior estimation. Moreover, unlike existing algorithms with double actor-critic frameworks, TDDR does not introduce any additional hyperparameters, significantly simplifying the design and implementation process. Experiments demonstrate that TDDR exhibits strong competitiveness compared to benchmark algorithms in challenging continuous control tasks.<|reference_end|>
|
arxiv
|
@article{chen2024double,
title={Double Actor-Critic with TD Error-Driven Regularization in Reinforcement
Learning},
author={Haohui Chen, Zhiyong Chen, Aoxiang Liu, and Wentuo Fang},
journal={arXiv preprint arXiv:2409.19231},
year={2024},
archivePrefix={arXiv},
eprint={2409.19231},
primaryClass={cs.LG cs.AI}
}
|
chen2024double
|
arxiv-663046
|
2409.19232
|
TrojVLM: Backdoor Attack Against Vision Language Models
|
<|reference_start|>TrojVLM: Backdoor Attack Against Vision Language Models: The emergence of Vision Language Models (VLMs) is a significant advancement in integrating computer vision with Large Language Models (LLMs) to produce detailed text descriptions based on visual inputs, yet it introduces new security vulnerabilities. Unlike prior work that centered on single modalities or classification tasks, this study introduces TrojVLM, the first exploration of backdoor attacks aimed at VLMs engaged in complex image-to-text generation. Specifically, TrojVLM inserts predetermined target text into output text when encountering poisoned images. Moreover, a novel semantic preserving loss is proposed to ensure the semantic integrity of the original image content. Our evaluation on image captioning and visual question answering (VQA) tasks confirms the effectiveness of TrojVLM in maintaining original semantic content while triggering specific target text outputs. This study not only uncovers a critical security risk in VLMs and image-to-text generation but also sets a foundation for future research on securing multimodal models against such sophisticated threats.<|reference_end|>
|
arxiv
|
@article{lyu2024trojvlm:,
title={TrojVLM: Backdoor Attack Against Vision Language Models},
author={Weimin Lyu, Lu Pang, Tengfei Ma, Haibin Ling, Chao Chen},
journal={arXiv preprint arXiv:2409.19232},
year={2024},
archivePrefix={arXiv},
eprint={2409.19232},
primaryClass={cs.CV}
}
|
lyu2024trojvlm:
|
arxiv-663047
|
2409.19234
|
Decoding Android Malware with a Fraction of Features: An Attention-Enhanced MLP-SVM Approach
|
<|reference_start|>Decoding Android Malware with a Fraction of Features: An Attention-Enhanced MLP-SVM Approach: The escalating sophistication of Android malware poses significant challenges to traditional detection methods, necessitating innovative approaches that can efficiently identify and classify threats with high precision. This paper introduces a novel framework that synergistically integrates an attention-enhanced Multi-Layer Perceptron (MLP) with a Support Vector Machine (SVM) to make Android malware detection and classification more effective. By carefully analyzing a mere 47 features out of over 9,760 available in the comprehensive CCCS-CIC-AndMal-2020 dataset, our MLP-SVM model achieves an impressive accuracy over 99% in identifying malicious applications. The MLP, enhanced with an attention mechanism, focuses on the most discriminative features and further reduces the 47 features to only 14 components using Linear Discriminant Analysis (LDA). Despite this significant reduction in dimensionality, the SVM component, equipped with an RBF kernel, excels in mapping these components to a high-dimensional space, facilitating precise classification of malware into their respective families. Rigorous evaluations, encompassing accuracy, precision, recall, and F1-score metrics, confirm the superiority of our approach compared to existing state-of-the-art techniques. The proposed framework not only significantly reduces the computational complexity by leveraging a compact feature set but also exhibits resilience against the evolving Android malware landscape.<|reference_end|>
|
arxiv
|
@article{hakim2024decoding,
title={Decoding Android Malware with a Fraction of Features: An
Attention-Enhanced MLP-SVM Approach},
author={Safayat Bin Hakim, Muhammad Adil, Kamal Acharya, Houbing Herbert Song},
journal={arXiv preprint arXiv:2409.19234},
year={2024},
archivePrefix={arXiv},
eprint={2409.19234},
primaryClass={cs.CR cs.LG}
}
|
hakim2024decoding
|
arxiv-663048
|
2409.19237
|
The Price of Pessimism for Automated Defense
|
<|reference_start|>The Price of Pessimism for Automated Defense: The well-worn George Box aphorism ``all models are wrong, but some are useful'' is particularly salient in the cybersecurity domain, where the assumptions built into a model can have substantial financial or even national security impacts. Computer scientists are often asked to optimize for worst-case outcomes, and since security is largely focused on risk mitigation, preparing for the worst-case scenario appears rational. In this work, we demonstrate that preparing for the worst case rather than the most probable case may yield suboptimal outcomes for learning agents. Through the lens of stochastic Bayesian games, we first explore different attacker knowledge modeling assumptions that impact the usefulness of models to cybersecurity practitioners. By considering different models of attacker knowledge about the state of the game and a defender's hidden information, we find that there is a cost to the defender for optimizing against the worst case.<|reference_end|>
|
arxiv
|
@article{galinkin2024the,
title={The Price of Pessimism for Automated Defense},
author={Erick Galinkin, Emmanouil Pountourakis, Spiros Mancoridis},
journal={arXiv preprint arXiv:2409.19237},
year={2024},
archivePrefix={arXiv},
eprint={2409.19237},
primaryClass={cs.CR cs.AI}
}
|
galinkin2024the
|
arxiv-663049
|
2409.19239
|
Zorro: A Flexible and Differentiable Parametric Family of Activation Functions That Extends ReLU and GELU
|
<|reference_start|>Zorro: A Flexible and Differentiable Parametric Family of Activation Functions That Extends ReLU and GELU: Even in recent neural network architectures such as Transformers and Extended LSTM (xLSTM), and traditional ones like Convolutional Neural Networks, Activation Functions are an integral part of nearly all neural networks. They enable more effective training and capture nonlinear data patterns. More than 400 functions have been proposed over the last 30 years, including fixed or trainable parameters, but only a few are widely used. ReLU is one of the most frequently used, with GELU and Swish variants increasingly appearing. However, ReLU presents non-differentiable points and exploding gradient issues, while testing different parameters of GELU and Swish variants produces varying results, needing more parameters to adapt to datasets and architectures. This article introduces a novel set of activation functions called Zorro, a continuously differentiable and flexible family comprising five main functions fusing ReLU and Sigmoid. Zorro functions are smooth and adaptable, and serve as information gates, aligning with ReLU in the 0-1 range, offering an alternative to ReLU without the need for normalization, neuron death, or gradient explosions. Zorro also approximates functions like Swish, GELU, and DGELU, providing parameters to adjust to different datasets and architectures. We tested it on fully connected, convolutional, and transformer architectures to demonstrate its effectiveness.<|reference_end|>
|
arxiv
|
@article{roodschild2024zorro:,
title={Zorro: A Flexible and Differentiable Parametric Family of Activation
Functions That Extends ReLU and GELU},
author={Matias Roodschild, Jorge Gotay-Sardi~nas, Victor A. Jimenez, Adrian
Will},
journal={arXiv preprint arXiv:2409.19239},
year={2024},
archivePrefix={arXiv},
eprint={2409.19239},
primaryClass={cs.LG cs.NE}
}
|
roodschild2024zorro:
|
arxiv-663050
|
2409.19242
|
SciDoc2Diagrammer-MAF: Towards Generation of Scientific Diagrams from Documents guided by Multi-Aspect Feedback Refinement
|
<|reference_start|>SciDoc2Diagrammer-MAF: Towards Generation of Scientific Diagrams from Documents guided by Multi-Aspect Feedback Refinement: Automating the creation of scientific diagrams from academic papers can significantly streamline the development of tutorials, presentations, and posters, thereby saving time and accelerating the process. Current text-to-image models struggle with generating accurate and visually appealing diagrams from long-context inputs. We propose SciDoc2Diagram, a task that extracts relevant information from scientific papers and generates diagrams, along with a benchmarking dataset, SciDoc2DiagramBench. We develop a multi-step pipeline SciDoc2Diagrammer that generates diagrams based on user intentions using intermediate code generation. We observed that initial diagram drafts were often incomplete or unfaithful to the source, leading us to develop SciDoc2Diagrammer-Multi-Aspect-Feedback (MAF), a refinement strategy that significantly enhances factual correctness and visual appeal and outperforms existing models on both automatic and human judgement.<|reference_end|>
|
arxiv
|
@article{mondal2024scidoc2diagrammer-maf:,
title={SciDoc2Diagrammer-MAF: Towards Generation of Scientific Diagrams from
Documents guided by Multi-Aspect Feedback Refinement},
author={Ishani Mondal, Zongxia Li, Yufang Hou, Anandhavelu Natarajan, Aparna
Garimella, Jordan Boyd-Graber},
journal={Empirical Methods in Natural Language Processing 2024},
year={2024},
archivePrefix={arXiv},
eprint={2409.19242},
primaryClass={cs.CL}
}
|
mondal2024scidoc2diagrammer-maf:
|
arxiv-663051
|
2409.19243
|
Jointly modelling the evolution of community structure and language in online extremist groups
|
<|reference_start|>Jointly modelling the evolution of community structure and language in online extremist groups: Group interactions take place within a particular socio-temporal context, which should be taken into account when modelling communities. We propose a method for jointly modelling community structure and language over time, and apply it in the context of extremist anti-women online groups (collectively known as the manosphere). Our model derives temporally grounded embeddings for words and users, which evolve over the training window. We show that this approach outperforms prior models which lacked one of these components (i.e. not incorporating social structure, or using static word embeddings). Using these embeddings, we investigate the evolution of users and words within these communities in three ways: (i) we model a user as a sequence of embeddings and forecast their affinity groups beyond the training window, (ii) we illustrate how word evolution is useful in the context of temporal events, and (iii) we characterise the propensity for violent language within subgroups of the manosphere.<|reference_end|>
|
arxiv
|
@article{de kock2024jointly,
title={Jointly modelling the evolution of community structure and language in
online extremist groups},
author={Christine de Kock},
journal={arXiv preprint arXiv:2409.19243},
year={2024},
archivePrefix={arXiv},
eprint={2409.19243},
primaryClass={cs.SI cs.CL}
}
|
de kock2024jointly
|
arxiv-663052
|
2409.19245
|
Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learning
|
<|reference_start|>Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learning: Online continual learning requires the models to learn from constant, endless streams of data. While significant efforts have been made in this field, most were focused on mitigating the catastrophic forgetting issue to achieve better classification ability, at the cost of a much heavier training workload. They overlooked that in real-world scenarios, e.g., in high-speed data stream environments, data do not pause to accommodate slow models. In this paper, we emphasize that model throughput -- defined as the maximum number of training samples that a model can process within a unit of time -- is equally important. It directly limits how much data a model can utilize and presents a challenging dilemma for current methods. With this understanding, we revisit key challenges in OCL from both empirical and theoretical perspectives, highlighting two critical issues beyond the well-documented catastrophic forgetting: Model's ignorance: the single-pass nature of OCL challenges models to learn effective features within constrained training time and storage capacity, leading to a trade-off between effective learning and model throughput; Model's myopia: the local learning nature of OCL on the current task leads the model to adopt overly simplified, task-specific features and excessively sparse classifier, resulting in the gap between the optimal solution for the current task and the global objective. To tackle these issues, we propose the Non-sparse Classifier Evolution framework (NsCE) to facilitate effective global discriminative feature learning with minimal time cost. NsCE integrates non-sparse maximum separation regularization and targeted experience replay techniques with the help of pre-trained models, enabling rapid acquisition of new globally discriminative features.<|reference_end|>
|
arxiv
|
@article{wang2024forgetting,,
title={Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online
Continual Learning},
author={Xinrui Wang, Chuanxing Geng, Wenhai Wan, Shao-yuan Li, Songcan Chen},
journal={arXiv preprint arXiv:2409.19245},
year={2024},
archivePrefix={arXiv},
eprint={2409.19245},
primaryClass={cs.LG}
}
|
wang2024forgetting,
|
arxiv-663053
|
2409.19247
|
Edit-Constrained Decoding for Sentence Simplification
|
<|reference_start|>Edit-Constrained Decoding for Sentence Simplification: We propose edit operation based lexically constrained decoding for sentence simplification. In sentence simplification, lexical paraphrasing is one of the primary procedures for rewriting complex sentences into simpler correspondences. While previous studies have confirmed the efficacy of lexically constrained decoding on this task, their constraints can be loose and may lead to sub-optimal generation. We address this problem by designing constraints that replicate the edit operations conducted in simplification and defining stricter satisfaction conditions. Our experiments indicate that the proposed method consistently outperforms the previous studies on three English simplification corpora commonly used in this task.<|reference_end|>
|
arxiv
|
@article{zetsu2024edit-constrained,
title={Edit-Constrained Decoding for Sentence Simplification},
author={Tatsuya Zetsu, Yuki Arase, Tomoyuki Kajiwara},
journal={arXiv preprint arXiv:2409.19247},
year={2024},
archivePrefix={arXiv},
eprint={2409.19247},
primaryClass={cs.CL cs.AI}
}
|
zetsu2024edit-constrained
|
arxiv-663054
|
2409.19248
|
Integrating Data Mining and Predictive Modeling Techniques for Enhanced Retail Optimization
|
<|reference_start|>Integrating Data Mining and Predictive Modeling Techniques for Enhanced Retail Optimization: Predictive modeling and time-pattern analysis are increasingly critical in this swiftly shifting retail environment to improve operational efficiency and informed decision-making. This paper reports a comprehensive application of state-of-the-art machine learning to the retailing domain with a specific focus on association rule mining, sequential pattern mining, and time-series forecasting. Association rules: Relationship Mining This provides the key product relationships and customer buying patterns that form the basis of individually tailored marketing campaigns. Sequential pattern mining: Using the PrefixSpan algorithm, it identifies frequent sequences of purchasing products-extremely powerful insights into consumer behavior and also better management of the inventories. What is applied for sales trend forecasting models Prophet applies on historical transaction data over seasonality, holidays, and long-term growth. The forecast results allow predicting demand variations, thus helping in proper inventory alignment and avoiding overstocking or understocking of inventory. Our results are checked through the help of metrics like MAE (Mean Absolute Error) and RMSE (Root Mean Squared Error) to ensure our predictions are strong and accurate. We will combine the aspects of all of these techniques to prove how predictive modeling and temporal pattern analysis can help optimize control over inventory, enhance marketing effectiveness, and position retail businesses as they rise to ever greater heights. This entire methodology demonstrates the flexibility with which data-driven strategies can be leveraged to revitalize traditional retailing practices.<|reference_end|>
|
arxiv
|
@article{m2024integrating,
title={Integrating Data Mining and Predictive Modeling Techniques for Enhanced
Retail Optimization},
author={Sri Darshan M, Jaisachin B, and NithinRaj N},
journal={IJCSIS (Volume 22 No. 5 ) 2024},
year={2024},
archivePrefix={arXiv},
eprint={2409.19248},
primaryClass={cs.CE}
}
|
m2024integrating
|
arxiv-663055
|
2409.19250
|
Fast and Accurate Task Planning using Neuro-Symbolic Language Models and Multi-level Goal Decomposition
|
<|reference_start|>Fast and Accurate Task Planning using Neuro-Symbolic Language Models and Multi-level Goal Decomposition: In robotic task planning, symbolic planners using rule-based representations like PDDL are effective but struggle with long-sequential tasks in complicated planning environments due to exponentially increasing search space. Recently, Large Language Models (LLMs) based on artificial neural networks have emerged as promising alternatives for autonomous robot task planning, offering faster inference and leveraging commonsense knowledge. However, they typically suffer from lower success rates. In this paper, to address the limitations of the current symbolic (slow speed) or LLM-based approaches (low accuracy), we propose a novel neuro-symbolic task planner that decomposes complex tasks into subgoals using LLM and carries out task planning for each subgoal using either symbolic or MCTS-based LLM planners, depending on the subgoal complexity. Generating subgoals helps reduce planning time and improve success rates by narrowing the overall search space and enabling LLMs to focus on smaller, more manageable tasks. Our method significantly reduces planning time while maintaining a competitive success rate, as demonstrated through experiments in different public task planning domains, as well as real-world and simulated robotics environments.<|reference_end|>
|
arxiv
|
@article{kwon2024fast,
title={Fast and Accurate Task Planning using Neuro-Symbolic Language Models and
Multi-level Goal Decomposition},
author={Minseo Kwon, Yaesol Kim and Young J. Kim},
journal={arXiv preprint arXiv:2409.19250},
year={2024},
archivePrefix={arXiv},
eprint={2409.19250},
primaryClass={cs.RO}
}
|
kwon2024fast
|
arxiv-663056
|
2409.19252
|
Beyond Euclidean: Dual-Space Representation Learning for Weakly Supervised Video Violence Detection
|
<|reference_start|>Beyond Euclidean: Dual-Space Representation Learning for Weakly Supervised Video Violence Detection: While numerous Video Violence Detection (VVD) methods have focused on representation learning in Euclidean space, they struggle to learn sufficiently discriminative features, leading to weaknesses in recognizing normal events that are visually similar to violent events (\emph{i.e.}, ambiguous violence). In contrast, hyperbolic representation learning, renowned for its ability to model hierarchical and complex relationships between events, has the potential to amplify the discrimination between visually similar events. Inspired by these, we develop a novel Dual-Space Representation Learning (DSRL) method for weakly supervised VVD to utilize the strength of both Euclidean and hyperbolic geometries, capturing the visual features of events while also exploring the intrinsic relations between events, thereby enhancing the discriminative capacity of the features. DSRL employs a novel information aggregation strategy to progressively learn event context in hyperbolic spaces, which selects aggregation nodes through layer-sensitive hyperbolic association degrees constrained by hyperbolic Dirichlet energy. Furthermore, DSRL attempts to break the cyber-balkanization of different spaces, utilizing cross-space attention to facilitate information interactions between Euclidean and hyperbolic space to capture better discriminative features for final violence detection. Comprehensive experiments demonstrate the effectiveness of our proposed DSRL.<|reference_end|>
|
arxiv
|
@article{leng2024beyond,
title={Beyond Euclidean: Dual-Space Representation Learning for Weakly
Supervised Video Violence Detection},
author={Jiaxu Leng, Zhanjie Wu, Mingpi Tan, Yiran Liu, Ji Gan, Haosheng Chen,
Xinbo Gao},
journal={arXiv preprint arXiv:2409.19252},
year={2024},
archivePrefix={arXiv},
eprint={2409.19252},
primaryClass={cs.CV}
}
|
leng2024beyond
|
arxiv-663057
|
2409.19255
|
DENEB: A Hallucination-Robust Automatic Evaluation Metric for Image Captioning
|
<|reference_start|>DENEB: A Hallucination-Robust Automatic Evaluation Metric for Image Captioning: In this work, we address the challenge of developing automatic evaluation metrics for image captioning, with a particular focus on robustness against hallucinations. Existing metrics are often inadequate for handling hallucinations, primarily due to their limited ability to compare candidate captions with multifaceted reference captions. To address this shortcoming, we propose DENEB, a novel supervised automatic evaluation metric specifically robust against hallucinations. DENEB incorporates the Sim-Vec Transformer, a mechanism that processes multiple references simultaneously, thereby efficiently capturing the similarity between an image, a candidate caption, and reference captions. To train DENEB, we construct the diverse and balanced Nebula dataset comprising 32,978 images, paired with human judgments provided by 805 annotators. We demonstrated that DENEB achieves state-of-the-art performance among existing LLM-free metrics on the FOIL, Composite, Flickr8K-Expert, Flickr8K-CF, Nebula, and PASCAL-50S datasets, validating its effectiveness and robustness against hallucinations.<|reference_end|>
|
arxiv
|
@article{matsuda2024deneb:,
title={DENEB: A Hallucination-Robust Automatic Evaluation Metric for Image
Captioning},
author={Kazuki Matsuda and Yuiga Wada and Komei Sugiura},
journal={arXiv preprint arXiv:2409.19255},
year={2024},
archivePrefix={arXiv},
eprint={2409.19255},
primaryClass={cs.CV cs.AI cs.CL}
}
|
matsuda2024deneb:
|
arxiv-663058
|
2409.19256
|
HybridFlow: A Flexible and Efficient RLHF Framework
|
<|reference_start|>HybridFlow: A Flexible and Efficient RLHF Framework: Reinforcement Learning from Human Feedback (RLHF) is widely used in Large Language Model (LLM) alignment. Traditional RL can be modeled as a dataflow, where each node represents computation of a neural network (NN) and each edge denotes data dependencies between the NNs. RLHF complicates the dataflow by expanding each node into a distributed LLM training or generation program, and each edge into a many-to-many multicast. Traditional RL frameworks execute the dataflow using a single controller to instruct both intra-node computation and inter-node communication, which can be inefficient in RLHF due to large control dispatch overhead for distributed intra-node computation. Existing RLHF systems adopt a multi-controller paradigm, which can be inflexible due to nesting distributed computation and data communication. We propose HybridFlow, which combines single-controller and multi-controller paradigms in a hybrid manner to enable flexible representation and efficient execution of the RLHF dataflow. We carefully design a set of hierarchical APIs that decouple and encapsulate computation and data dependencies in the complex RLHF dataflow, allowing efficient operation orchestration to implement RLHF algorithms and flexible mapping of the computation onto various devices. We further design a 3D-HybridEngine for efficient actor model resharding between training and generation phases, with zero memory redundancy and significantly reduced communication overhead. Our experimental results demonstrate 1.53$\times$~20.57$\times$ throughput improvement when running various RLHF algorithms using HybridFlow, as compared with state-of-the-art baselines. HybridFlow source code will be available at https://github.com/volcengine/verl.<|reference_end|>
|
arxiv
|
@article{sheng2024hybridflow:,
title={HybridFlow: A Flexible and Efficient RLHF Framework},
author={Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru
Zhang, Yanghua Peng, Haibin Lin, Chuan Wu},
journal={arXiv preprint arXiv:2409.19256},
year={2024},
doi={10.1145/3689031.3696075},
archivePrefix={arXiv},
eprint={2409.19256},
primaryClass={cs.LG cs.DC}
}
|
sheng2024hybridflow:
|
arxiv-663059
|
2409.19257
|
LISTN: Lexicon induction with socio-temporal nuance
|
<|reference_start|>LISTN: Lexicon induction with socio-temporal nuance: Research on extremist online communities frequently utilizes linguistic analysis to explore group dynamics and behaviour. Existing studies often rely on outdated lexicons that do not capture the evolving nature of in-group language, nor the social structure of the community. This paper proposes a novel method for inducing in-group lexicons which incorporates its socio-temporal context. Using dynamic word and user embeddings trained on conversations from online anti-women communities, our approach outperforms prior methods for lexicon induction. We provide a new lexicon of manosphere terms, validated by human experts, which quantifies the relevance of each term to a specific sub-community. We present novel insights on in-group language which illustrate the utility of this approach.<|reference_end|>
|
arxiv
|
@article{de kock2024listn:,
title={LISTN: Lexicon induction with socio-temporal nuance},
author={Christine de Kock},
journal={arXiv preprint arXiv:2409.19257},
year={2024},
archivePrefix={arXiv},
eprint={2409.19257},
primaryClass={cs.CL cs.SI}
}
|
de kock2024listn:
|
arxiv-663060
|
2409.19258
|
VecLSTM: Trajectory Data Processing and Management for Activity Recognition through LSTM Vectorization and Database Integration
|
<|reference_start|>VecLSTM: Trajectory Data Processing and Management for Activity Recognition through LSTM Vectorization and Database Integration: Activity recognition is a challenging task due to the large scale of trajectory data and the need for prompt and efficient processing. Existing methods have attempted to mitigate this problem by employing traditional LSTM architectures, but these approaches often suffer from inefficiencies in processing large datasets. In response to this challenge, we propose VecLSTM, a novel framework that enhances the performance and efficiency of LSTM-based neural networks. Unlike conventional approaches, VecLSTM incorporates vectorization layers, leveraging optimized mathematical operations to process input sequences more efficiently. We have implemented VecLSTM and incorporated it into the MySQL database. To evaluate the effectiveness of VecLSTM, we compare its performance against a conventional LSTM model using a dataset comprising 1,467,652 samples with seven unique labels. Experimental results demonstrate superior accuracy and efficiency compared to the state-of-the-art, with VecLSTM achieving a validation accuracy of 85.57\%, a test accuracy of 85.47\%, and a weighted F1-score of 0.86. Furthermore, VecLSTM significantly reduces training time, offering a 26.2\% reduction compared to traditional LSTM models.<|reference_end|>
|
arxiv
|
@article{monir2024veclstm:,
title={VecLSTM: Trajectory Data Processing and Management for Activity
Recognition through LSTM Vectorization and Database Integration},
author={Solmaz Seyed Monir, Dongfang Zhao},
journal={arXiv preprint arXiv:2409.19258},
year={2024},
archivePrefix={arXiv},
eprint={2409.19258},
primaryClass={cs.LG cs.AI cs.DB cs.NE}
}
|
monir2024veclstm:
|
arxiv-663061
|
2409.19262
|
An Efficient Multi-threaded Collaborative Filtering Approach in Recommendation System
|
<|reference_start|>An Efficient Multi-threaded Collaborative Filtering Approach in Recommendation System: Recommender systems are a subset of information filtering systems designed to predict and suggest items that users may find interesting or relevant based on their preferences, behaviors, or interactions. By analyzing user data such as past activities, ratings, and preferences, these systems generate personalized recommendations for products, services, or content, with common applications including online retail, media streaming platforms, and social media. Recommender systems are typically categorized into three types: content-based filtering, which recommends items similar to those the user has shown interest in; collaborative filtering, which analyzes the preferences of similar users; and hybrid methods, which combine both approaches to improve accuracy. These systems enhance user experience by reducing information overload and providing personalized suggestions, thus increasing engagement and satisfaction. However, building a scalable recommendation system capable of handling numerous users efficiently is a significant challenge, particularly when considering both performance consistency and user data security, which are emerging research topics. The primary objective of this research is to address these challenges by reducing the processing time in recommendation systems. A multithreaded similarity approach is employed to achieve this, where users are divided into independent threads that run in parallel. This parallelization significantly reduces computation time compared to traditional methods, resulting in a faster, more efficient, and scalable recommendation system that ensures improved performance without compromising user data security.<|reference_end|>
|
arxiv
|
@article{hasan2024an,
title={An Efficient Multi-threaded Collaborative Filtering Approach in
Recommendation System},
author={Mahamudul Hasan},
journal={arXiv preprint arXiv:2409.19262},
year={2024},
archivePrefix={arXiv},
eprint={2409.19262},
primaryClass={cs.IR}
}
|
hasan2024an
|
arxiv-663062
|
2409.19267
|
Utilizing Collaborative Filtering in a Personalized Research-Paper Recommendation System
|
<|reference_start|>Utilizing Collaborative Filtering in a Personalized Research-Paper Recommendation System: Recommendation system is such a platform that helps people to easily find out the things they need within a few seconds. It is implemented based on the preferences of similar users or items. In this digital era, the internet has provided us with huge opportunities to use a lot of open resources for our own needs. But there are too many resources on the internet from which finding the precise one is a difficult job. Recommendation system has made this easier for people. Research-paper recommendation system is a system that is developed for people with common research interests using a collaborative filtering recommender system. In this paper, coauthor, keyword, reference, and common citation similarities are calculated using Jaccard Similarity to find the final similarity and to find the top-n similar users. Based on the test of top-n similar users of the target user research paper recommendations have been made. Finally, the accuracy of our recommendation system has been calculated. An impressive result has been found using our proposed system.<|reference_end|>
|
arxiv
|
@article{hasan2024utilizing,
title={Utilizing Collaborative Filtering in a Personalized Research-Paper
Recommendation System},
author={Mahamudul Hasan, Anika Tasnim Islam, Nabila Islam},
journal={arXiv preprint arXiv:2409.19267},
year={2024},
archivePrefix={arXiv},
eprint={2409.19267},
primaryClass={cs.IR}
}
|
hasan2024utilizing
|
arxiv-663063
|
2409.19269
|
PDCFNet: Enhancing Underwater Images through Pixel Difference Convolution
|
<|reference_start|>PDCFNet: Enhancing Underwater Images through Pixel Difference Convolution: Majority of deep learning methods utilize vanilla convolution for enhancing underwater images. While vanilla convolution excels in capturing local features and learning the spatial hierarchical structure of images, it tends to smooth input images, which can somewhat limit feature expression and modeling. A prominent characteristic of underwater degraded images is blur, and the goal of enhancement is to make the textures and details (high-frequency features) in the images more visible. Therefore, we believe that leveraging high-frequency features can improve enhancement performance. To address this, we introduce Pixel Difference Convolution (PDC), which focuses on gradient information with significant changes in the image, thereby improving the modeling of enhanced images. We propose an underwater image enhancement network, PDCFNet, based on PDC and cross-level feature fusion. Specifically, we design a detail enhancement module based on PDC that employs parallel PDCs to capture high-frequency features, leading to better detail and texture enhancement. The designed cross-level feature fusion module performs operations such as concatenation and multiplication on features from different levels, ensuring sufficient interaction and enhancement between diverse features. Our proposed PDCFNet achieves a PSNR of 27.37 and an SSIM of 92.02 on the UIEB dataset, attaining the best performance to date. Our code is available at https://github.com/zhangsong1213/PDCFNet.<|reference_end|>
|
arxiv
|
@article{zhang2024pdcfnet:,
title={PDCFNet: Enhancing Underwater Images through Pixel Difference
Convolution},
author={Song Zhang, Daoliang Li, Ran Zhao},
journal={arXiv preprint arXiv:2409.19269},
year={2024},
archivePrefix={arXiv},
eprint={2409.19269},
primaryClass={cs.CV}
}
|
zhang2024pdcfnet:
|
arxiv-663064
|
2409.19270
|
OpenSep: Leveraging Large Language Models with Textual Inversion for Open World Audio Separation
|
<|reference_start|>OpenSep: Leveraging Large Language Models with Textual Inversion for Open World Audio Separation: Audio separation in real-world scenarios, where mixtures contain a variable number of sources, presents significant challenges due to limitations of existing models, such as over-separation, under-separation, and dependence on predefined training sources. We propose OpenSep, a novel framework that leverages large language models (LLMs) for automated audio separation, eliminating the need for manual intervention and overcoming source limitations. OpenSep uses textual inversion to generate captions from audio mixtures with off-the-shelf audio captioning models, effectively parsing the sound sources present. It then employs few-shot LLM prompting to extract detailed audio properties of each parsed source, facilitating separation in unseen mixtures. Additionally, we introduce a multi-level extension of the mix-and-separate training framework to enhance modality alignment by separating single source sounds and mixtures simultaneously. Extensive experiments demonstrate OpenSep's superiority in precisely separating new, unseen, and variable sources in challenging mixtures, outperforming SOTA baseline methods. Code is released at https://github.com/tanvir-utexas/OpenSep.git<|reference_end|>
|
arxiv
|
@article{mahmud2024opensep:,
title={OpenSep: Leveraging Large Language Models with Textual Inversion for
Open World Audio Separation},
author={Tanvir Mahmud and Diana Marculescu},
journal={arXiv preprint arXiv:2409.19270},
year={2024},
archivePrefix={arXiv},
eprint={2409.19270},
primaryClass={cs.SD cs.AI eess.AS}
}
|
mahmud2024opensep:
|
arxiv-663065
|
2409.19272
|
Perception Compressor:A training-free prompt compression method in long context scenarios
|
<|reference_start|>Perception Compressor:A training-free prompt compression method in long context scenarios: Large Language Models (LLMs) demonstrate exceptional capabilities in various scenarios. However, they suffer from much redundant information and tend to be lost in the middle in long context scenarios, leading to inferior performance. To address these challenges, we present Perception Compressor, a training-free prompt compression method. It includes a dual-slope ratio allocator to dynamically assign compression ratios and open-book ratios, a perception retriever that leverages guiding questions and instruction to retrieve the most relevant demonstrations, and a semi-guided iterative compression that retains key information at the token level while removing tokens that distract the LLM. We conduct extensive experiments on long context benchmarks, i.e., NaturalQuestions, LongBench, and MuSiQue. Experiment results show that Perception Compressor outperforms existing methods by a large margin, achieving state-of-the-art performance.<|reference_end|>
|
arxiv
|
@article{tang2024perception,
title={Perception Compressor:A training-free prompt compression method in long
context scenarios},
author={Jiwei Tang, Jin Xu, Tingwei Lu, Zhicheng Zhang, Yiming Zhao, Lin Hai,
Hai-Tao Zheng},
journal={arXiv preprint arXiv:2409.19272},
year={2024},
archivePrefix={arXiv},
eprint={2409.19272},
primaryClass={cs.CL}
}
|
tang2024perception
|
arxiv-663066
|
2409.19275
|
Implicit Euler Discrete-Time Set-Valued Admittance Control for Impact-Contact Force Control
|
<|reference_start|>Implicit Euler Discrete-Time Set-Valued Admittance Control for Impact-Contact Force Control: Admittance control is a commonly used strategy for regulating robotic systems, such as quadruped and humanoid robots, allowing them to respond compliantly to contact forces during interactions with their environments. However, it can lead to instability and unsafe behaviors like snapping back and overshooting due to torque saturation from impacts with unknown stiffness environments. This paper introduces a novel admittance controller that ensures stable force control after impacting unknown stiffness environments by leveraging the differentiability of impact-contact forces. The controller is mathematically represented by a differential algebraic inclusion (DAI) comprising two interdependent set-valued loops. The first loop employs set-valued first-order sliding mode control (SMC) to limit input torque post-impact. The second loop utilizes the multivariable super-twisting algorithm (MSTA) to mitigate unstable motion caused by impact forces when interacting with unknown stiffness environments. Implementing this proposed admittance control in digital settings presents challenges due to the interconnected structure of the two set-valued loops, unlike implicit Euler discretization methods for set-valued SMCs. To facilitate implementation, this paper offers a new algorithm for implicit Euler discretization of the DAI. Simulation and experimental results demonstrate that the proposed admittance controller outperforms state-of-the-art methods.<|reference_end|>
|
arxiv
|
@article{li2024implicit,
title={Implicit Euler Discrete-Time Set-Valued Admittance Control for
Impact-Contact Force Control},
author={Ke Li, Xiaogang Xiong, Anjia Wang, Ying Qu, and Yunjiang Lou},
journal={arXiv preprint arXiv:2409.19275},
year={2024},
archivePrefix={arXiv},
eprint={2409.19275},
primaryClass={eess.SY cs.SY}
}
|
li2024implicit
|
arxiv-663067
|
2409.19277
|
Symmetry Preservation in Swarms of Oblivious Robots with Limited Visibility
|
<|reference_start|>Symmetry Preservation in Swarms of Oblivious Robots with Limited Visibility: In the general pattern formation (GPF) problem, a swarm of simple autonomous, disoriented robots must form a given pattern. The robots' simplicity imply a strong limitation: When the initial configuration is rotationally symmetric, only patterns with a similar symmetry can be formed [Yamashita, Suzyuki; TCS 2010]. The only known algorithm to form large patterns with limited visibility and without memory requires the robots to start in a near-gathering (a swarm of constant diameter) [Hahn et al.; SAND 2024]. However, not only do we not know any near-gathering algorithm guaranteed to preserve symmetry but most natural gathering strategies trivially increase symmetries [Castenow et al.; OPODIS 2022]. Thus, we study near-gathering without changing the swarm's rotational symmetry for disoriented, oblivious robots with limited visibility (the OBLOT-model, see [Flocchini et al.; 2019]). We introduce a technique based on the theory of dynamical systems to analyze how a given algorithm affects symmetry and provide sufficient conditions for symmetry preservation. Until now, it was unknown whether the considered OBLOT-model allows for any non-trivial algorithm that always preserves symmetry. Our first result shows that a variant of Go-to-the-Average always preserves symmetry but may sometimes lead to multiple, unconnected near-gathering clusters. Our second result is a symmetry-preserving near-gathering algorithm that works on swarms with a convex boundary (the outer boundary of the unit disc graph) and without holes (circles of diameter 1 inside the boundary without any robots).<|reference_end|>
|
arxiv
|
@article{gerlach2024symmetry,
title={Symmetry Preservation in Swarms of Oblivious Robots with Limited
Visibility},
author={Raphael Gerlach and S"oren von der Gracht and Christopher Hahn and
Jonas Harbig and Peter Kling},
journal={arXiv preprint arXiv:2409.19277},
year={2024},
archivePrefix={arXiv},
eprint={2409.19277},
primaryClass={cs.RO cs.DS}
}
|
gerlach2024symmetry
|
arxiv-663068
|
2409.19278
|
Explicit construction of recurrent neural networks effectively approximating discrete dynamical systems
|
<|reference_start|>Explicit construction of recurrent neural networks effectively approximating discrete dynamical systems: We consider arbitrary bounded discrete time series originating from dynamical system with recursivity. More precisely, we provide an explicit construction of recurrent neural networks which effectively approximate the corresponding discrete dynamical systems.<|reference_end|>
|
arxiv
|
@article{nakayama2024explicit,
title={Explicit construction of recurrent neural networks effectively
approximating discrete dynamical systems},
author={Chikara Nakayama and Tsuyoshi Yoneda},
journal={arXiv preprint arXiv:2409.19278},
year={2024},
archivePrefix={arXiv},
eprint={2409.19278},
primaryClass={cs.LG math.DS}
}
|
nakayama2024explicit
|
arxiv-663069
|
2409.19279
|
Distributed Optimization via Energy Conservation Laws in Dilated Coordinates
|
<|reference_start|>Distributed Optimization via Energy Conservation Laws in Dilated Coordinates: Optimizing problems in a distributed manner is critical for systems involving multiple agents with private data. Despite substantial interest, a unified method for analyzing the convergence rates of distributed optimization algorithms is lacking. This paper introduces an energy conservation approach for analyzing continuous-time dynamical systems in dilated coordinates. Instead of directly analyzing dynamics in the original coordinate system, we establish a conserved quantity, akin to physical energy, in the dilated coordinate system. Consequently, convergence rates can be explicitly expressed in terms of the inverse time-dilation factor. Leveraging this generalized approach, we formulate a novel second-order distributed accelerated gradient flow with a convergence rate of $O\left(1/t^{2-\epsilon}\right)$ in time $t$ for $\epsilon>0$. We then employ a semi second-order symplectic Euler discretization to derive a rate-matching algorithm with a convergence rate of $O\left(1/k^{2-\epsilon}\right)$ in $k$ iterations. To the best of our knowledge, this represents the most favorable convergence rate for any distributed optimization algorithm designed for smooth convex optimization. Its accelerated convergence behavior is benchmarked against various state-of-the-art distributed optimization algorithms on practical, large-scale problems.<|reference_end|>
|
arxiv
|
@article{baranwal2024distributed,
title={Distributed Optimization via Energy Conservation Laws in Dilated
Coordinates},
author={Mayank Baranwal, Kushal Chakrabarti},
journal={arXiv preprint arXiv:2409.19279},
year={2024},
archivePrefix={arXiv},
eprint={2409.19279},
primaryClass={math.OC cs.AI cs.LG cs.SY eess.SY math.DS}
}
|
baranwal2024distributed
|
arxiv-663070
|
2409.19281
|
Gesture Recognition for Feedback Based Mixed Reality and Robotic Fabrication: A Case Study of the UnLog Tower
|
<|reference_start|>Gesture Recognition for Feedback Based Mixed Reality and Robotic Fabrication: A Case Study of the UnLog Tower: Mixed Reality (MR) platforms enable users to interact with three-dimensional holographic instructions during the assembly and fabrication of highly custom and parametric architectural constructions without the necessity of two-dimensional drawings. Previous MR fabrication projects have primarily relied on digital menus and custom buttons as the interface for user interaction with the MR environment. Despite this approach being widely adopted, it is limited in its ability to allow for direct human interaction with physical objects to modify fabrication instructions within the MR environment. This research integrates user interactions with physical objects through real-time gesture recognition as input to modify, update or generate new digital information enabling reciprocal stimuli between the physical and the virtual environment. Consequently, the digital environment is generative of the user's provided interaction with physical objects to allow seamless feedback in the fabrication process. This research investigates gesture recognition for feedback-based MR workflows for robotic fabrication, human assembly, and quality control in the construction of the UnLog Tower.<|reference_end|>
|
arxiv
|
@article{kyaw2024gesture,
title={Gesture Recognition for Feedback Based Mixed Reality and Robotic
Fabrication: A Case Study of the UnLog Tower},
author={Alexander Htet Kyaw, Lawson Spencer, Sasa Zivkovic, Leslie Lok},
journal={arXiv preprint arXiv:2409.19281},
year={2024},
doi={10.1007/978-981-99-8405-3_28},
archivePrefix={arXiv},
eprint={2409.19281},
primaryClass={cs.HC cs.ET cs.RO}
}
|
kyaw2024gesture
|
arxiv-663071
|
2409.19283
|
Analyzing and Mitigating Inconsistency in Discrete Audio Tokens for Neural Codec Language Models
|
<|reference_start|>Analyzing and Mitigating Inconsistency in Discrete Audio Tokens for Neural Codec Language Models: Building upon advancements in Large Language Models (LLMs), the field of audio processing has seen increased interest in training audio generation tasks with discrete audio token sequences. However, directly discretizing audio by neural audio codecs often results in sequences that fundamentally differ from text sequences. Unlike text, where text token sequences are deterministic, discrete audio tokens can exhibit significant variability based on contextual factors, while still producing perceptually identical audio segments. We refer to this phenomenon as \textbf{Discrete Representation Inconsistency (DRI)}. This inconsistency can lead to a single audio segment being represented by multiple divergent sequences, which creates confusion in neural codec language models and results in omissions and repetitions during speech generation. In this paper, we quantitatively analyze the DRI phenomenon within popular audio tokenizers such as EnCodec. Our approach effectively mitigates the DRI phenomenon of the neural audio codec. Furthermore, extensive experiments on the neural codec language model over LibriTTS and large-scale MLS datases (44,000 hours) demonstrate the effectiveness and generality of our method. The demo of audio samples is available online~\footnote{\url{https://consistencyinneuralcodec.github.io}}.<|reference_end|>
|
arxiv
|
@article{liu2024analyzing,
title={Analyzing and Mitigating Inconsistency in Discrete Audio Tokens for
Neural Codec Language Models},
author={Wenrui Liu, Zhifang Guo, Jin Xu, Yuanjun Lv, Yunfei Chu, Zhou Zhao,
Junyang Lin},
journal={arXiv preprint arXiv:2409.19283},
year={2024},
archivePrefix={arXiv},
eprint={2409.19283},
primaryClass={eess.AS cs.SD}
}
|
liu2024analyzing
|
arxiv-663072
|
2409.19286
|
IM: Optimizing Byzantine Consensus for High-Performance Distributed Networks
|
<|reference_start|>IM: Optimizing Byzantine Consensus for High-Performance Distributed Networks: Byzantine Fault Tolerant (BFT) consensus, a crucial component of blockchains, has made significant advancements. However, the efficiency of existing protocols can still be damaged by certain attacks from faulty nodes and network instability. In this paper, we propose a novel Shared Mempool (SMP) protocol, namely IM, that enhances performance under these attacks. Technically, IM organizing microblocks into chains, combined with coding techniques, achieves totality and availability efficiently. IM can be easily integrated into a BFT protocol. We take Fast-HotStuff as an example and obtain the IM-FHS with guarantees of \emph{order keeping}, \emph{bandwidth adaptability} and \emph{over-distribution resistance}. IM-FHS is conducted in a system with up to 256 nodes, and experimental results validate the efficiency of our approach. IM-FHS achieves higher throughput and smaller latency with faulty nodes than Stratus-FHS, the state-of-the-art protocol, and the throughput gain increases as the number of fault nodes. In a system with 100 nodes with 33 faulty nodes, IM-FHS achieves 9 times the throughput of Stratus-FHS while maintaining 1/10 the latency when dealing with maximum resilience against faulty nodes.<|reference_end|>
|
arxiv
|
@article{zeng2024im:,
title={IM: Optimizing Byzantine Consensus for High-Performance Distributed
Networks},
author={Qingming Zeng (Harbin Institute of Technology, Shenzhen), Mo Li (The
Chinese University of Hongkong, Shenzhen), Ximing Fu (Harbin Institute of
Technology, Shenzhen), Chuanyi Liu (Harbin Institute of Technology, Shenzhen,
Peng Cheng Laboratory, Shenzhen), Hui Jiang (Tsinghua University, Baidu Inc)},
journal={arXiv preprint arXiv:2409.19286},
year={2024},
archivePrefix={arXiv},
eprint={2409.19286},
primaryClass={cs.DC}
}
|
zeng2024im:
|
arxiv-663073
|
2409.19289
|
FINE: Factorizing Knowledge for Initialization of Variable-sized Diffusion Models
|
<|reference_start|>FINE: Factorizing Knowledge for Initialization of Variable-sized Diffusion Models: Diffusion models often face slow convergence, and existing efficient training techniques, such as Parameter-Efficient Fine-Tuning (PEFT), are primarily designed for fine-tuning pre-trained models. However, these methods are limited in adapting models to variable sizes for real-world deployment, where no corresponding pre-trained models exist. To address this, we introduce FINE, a method based on the Learngene framework, to initializing downstream networks leveraging pre-trained models, while considering both model sizes and task-specific requirements. FINE decomposes pre-trained knowledge into the product of matrices (i.e., $U$, $\Sigma$, and $V$), where $U$ and $V$ are shared across network blocks as ``learngenes'', and $\Sigma$ remains layer-specific. During initialization, FINE trains only $\Sigma$ using a small subset of data, while keeping the learngene parameters fixed, marking it the first approach to integrate both size and task considerations in initialization. We provide a comprehensive benchmark for learngene-based methods in image generation tasks, and extensive experiments demonstrate that FINE consistently outperforms direct pre-training, particularly for smaller models, achieving state-of-the-art results across variable model sizes. FINE also offers significant computational and storage savings, reducing training steps by approximately $3N\times$ and storage by $5\times$, where $N$ is the number of models. Additionally, FINE's adaptability to tasks yields an average performance improvement of 4.29 and 3.30 in FID and sFID across multiple downstream datasets, highlighting its versatility and efficiency.<|reference_end|>
|
arxiv
|
@article{xie2024fine:,
title={FINE: Factorizing Knowledge for Initialization of Variable-sized
Diffusion Models},
author={Yucheng Xie, Fu Feng, Ruixiao Shi, Jing Wang, Xin Geng},
journal={arXiv preprint arXiv:2409.19289},
year={2024},
archivePrefix={arXiv},
eprint={2409.19289},
primaryClass={cs.CV}
}
|
xie2024fine:
|
arxiv-663074
|
2409.19291
|
CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling
|
<|reference_start|>CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling: In recent years, Contrastive Language-Image Pre-training (CLIP) has become a cornerstone in multimodal intelligence. However, recent studies have identified that the information loss in the CLIP encoding process is substantial, and CLIP tends to capture only coarse-grained features from the input. This deficiency significantly limits the ability of a single CLIP model to handle images rich in visual detail. In this work, we propose a simple yet effective model-agnostic strategy, Diversified Multiplet Upcycling (DMU), for CLIP. DMU efficiently fine-tunes a series of CLIP models that capture different feature spaces, from a dense pre-trained CLIP checkpoint, sharing parameters except for the Feed-Forward Network (FFN). These models can then be transformed into a CLIP-MoE with a larger model capacity, leading to significantly enhanced performance with minimal computational overhead. To the best of our knowledge, Diversified Multiplet Upcycling is the first approach to introduce sparsely activated MoE into CLIP foundation models. Extensive experiments demonstrate the significant performance of CLIP-MoE across various zero-shot retrieval, zero-shot image classification tasks, and downstream Multimodal Large Language Model (MLLM) benchmarks by serving as a vision encoder. Furthermore, Diversified Multiplet Upcycling enables the conversion of any dense CLIP model into CLIP-MoEs, which can seamlessly replace CLIP in a plug-and-play manner without requiring further adaptation in downstream frameworks. Through Diversified Multiplet Upcycling, we aim to provide valuable insights for future research on developing more efficient and effective multimodal learning systems.<|reference_end|>
|
arxiv
|
@article{zhang2024clip-moe:,
title={CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified
Multiplet Upcycling},
author={Jihai Zhang, Xiaoye Qu, Tong Zhu, Yu Cheng},
journal={arXiv preprint arXiv:2409.19291},
year={2024},
archivePrefix={arXiv},
eprint={2409.19291},
primaryClass={cs.CV cs.AI}
}
|
zhang2024clip-moe:
|
arxiv-663075
|
2409.19292
|
Fast Approximate Counting of Cycles
|
<|reference_start|>Fast Approximate Counting of Cycles: We consider the problem of approximate counting of triangles and longer fixed length cycles in directed graphs. For triangles, T\v{e}tek [ICALP'22] gave an algorithm that returns a $(1 \pm \eps)$-approximation in $\tilde{O}(n^\omega/t^{\omega-2})$ time, where $t$ is the unknown number of triangles in the given $n$ node graph and $\omega<2.372$ is the matrix multiplication exponent. We obtain an improved algorithm whose running time is, within polylogarithmic factors the same as that for multiplying an $n\times n/t$ matrix by an $n/t \times n$ matrix. We then extend our framework to obtain the first nontrivial $(1 \pm \eps)$-approximation algorithms for the number of $h$-cycles in a graph, for any constant $h\geq 3$. Our running time is \[\tilde{O}(\mathsf{MM}(n,n/t^{1/(h-2)},n)), \textrm{the time to multiply } n\times \frac{n}{t^{1/(h-2)}} \textrm{ by } \frac{n}{t^{1/(h-2)}}\times n \textrm{ matrices}.\] Finally, we show that under popular fine-grained hypotheses, this running time is optimal.<|reference_end|>
|
arxiv
|
@article{censor-hillel2024fast,
title={Fast Approximate Counting of Cycles},
author={Keren Censor-Hillel, Tomer Even and Virginia Vassilevska Williams},
journal={arXiv preprint arXiv:2409.19292},
year={2024},
archivePrefix={arXiv},
eprint={2409.19292},
primaryClass={cs.DS}
}
|
censor-hillel2024fast
|
arxiv-663076
|
2409.19293
|
VLAD-BuFF: Burst-aware Fast Feature Aggregation for Visual Place Recognition
|
<|reference_start|>VLAD-BuFF: Burst-aware Fast Feature Aggregation for Visual Place Recognition: Visual Place Recognition (VPR) is a crucial component of many visual localization pipelines for embodied agents. VPR is often formulated as an image retrieval task aimed at jointly learning local features and an aggregation method. The current state-of-the-art VPR methods rely on VLAD aggregation, which can be trained to learn a weighted contribution of features through their soft assignment to cluster centers. However, this process has two key limitations. Firstly, the feature-to-cluster weighting does not account for over-represented repetitive structures within a cluster, e.g., shadows or window panes; this phenomenon is also referred to as the `burstiness' problem, classically solved by discounting repetitive features before aggregation. Secondly, feature to cluster comparisons are compute-intensive for state-of-the-art image encoders with high-dimensional local features. This paper addresses these limitations by introducing VLAD-BuFF with two novel contributions: i) a self-similarity based feature discounting mechanism to learn Burst-aware features within end-to-end VPR training, and ii) Fast Feature aggregation by reducing local feature dimensions specifically through PCA-initialized learnable pre-projection. We benchmark our method on 9 public datasets, where VLAD-BuFF sets a new state of the art. Our method is able to maintain its high recall even for 12x reduced local feature dimensions, thus enabling fast feature aggregation without compromising on recall. Through additional qualitative studies, we show how our proposed weighting method effectively downweights the non-distinctive features. Source code: https://github.com/Ahmedest61/VLAD-BuFF/.<|reference_end|>
|
arxiv
|
@article{khaliq2024vlad-buff:,
title={VLAD-BuFF: Burst-aware Fast Feature Aggregation for Visual Place
Recognition},
author={Ahmad Khaliq, Ming Xu, Stephen Hausler, Michael Milford, Sourav Garg},
journal={arXiv preprint arXiv:2409.19293},
year={2024},
archivePrefix={arXiv},
eprint={2409.19293},
primaryClass={cs.CV}
}
|
khaliq2024vlad-buff:
|
arxiv-663077
|
2409.19298
|
Proceedings 13th International Workshop on Developments in Computational Models
|
<|reference_start|>Proceedings 13th International Workshop on Developments in Computational Models: This volume contains the proceedings of DCM 2023, the 13th International Workshop on Developments in Computational Models held on 2 July 2023 in Rome, Italy. DCM 2023 was organised as a one-day satellite event of FSCD 2023, the 8th International Conference on Formal Structures for Computation and Deduction. The aim of this workshop is to bring together researchers who are currently developing new computation models or new features for traditional computation models, in order to foster their interaction, to provide a forum for presenting new ideas and work in progress, and to enable newcomers to learn about current activities in this area.<|reference_end|>
|
arxiv
|
@article{alves2024proceedings,
title={Proceedings 13th International Workshop on Developments in Computational
Models},
author={Sandra Alves (University of Porto), Ian Mackie (London South Bank
University)},
journal={EPTCS 408, 2024},
year={2024},
doi={10.4204/EPTCS.408},
archivePrefix={arXiv},
eprint={2409.19298},
primaryClass={cs.LO cs.PL cs.SC}
}
|
alves2024proceedings
|
arxiv-663078
|
2409.19300
|
Sustaining model performance for covid-19 detection from dynamic audio data: Development and evaluation of a comprehensive drift-adaptive framework
|
<|reference_start|>Sustaining model performance for covid-19 detection from dynamic audio data: Development and evaluation of a comprehensive drift-adaptive framework: Background: The COVID-19 pandemic has highlighted the need for robust diagnostic tools capable of detecting the disease from diverse and evolving data sources. Machine learning models, especially convolutional neural networks (CNNs), have shown promise. However, the dynamic nature of real-world data can lead to model drift, where performance degrades over time as the underlying data distribution changes. Addressing this challenge is crucial to maintaining accuracy and reliability in diagnostic applications. Objective: This study aims to develop a framework that monitors model drift and employs adaptation mechanisms to mitigate performance fluctuations in COVID-19 detection models trained on dynamic audio data. Methods: Two crowd-sourced COVID-19 audio datasets, COVID-19 Sounds and COSWARA, were used. Each was divided into development and post-development periods. A baseline CNN model was trained and evaluated using cough recordings from the development period. Maximum mean discrepancy (MMD) was used to detect changes in data distributions and model performance between periods. Upon detecting drift, retraining was triggered to update the baseline model. Two adaptation approaches were compared: unsupervised domain adaptation (UDA) and active learning (AL). Results: UDA improved balanced accuracy by up to 22% and 24% for the COVID-19 Sounds and COSWARA datasets, respectively. AL yielded even greater improvements, with increases of up to 30% and 60%, respectively. Conclusions: The proposed framework addresses model drift in COVID-19 detection, enabling continuous adaptation to evolving data. This approach ensures sustained model performance, contributing to robust diagnostic tools for COVID-19 and potentially other infectious diseases.<|reference_end|>
|
arxiv
|
@article{ganitidis2024sustaining,
title={Sustaining model performance for covid-19 detection from dynamic audio
data: Development and evaluation of a comprehensive drift-adaptive framework},
author={Theofanis Ganitidis, Maria Athanasiou, Konstantinos Mitsis, Konstantia
Zarkogianni, Konstantina S. Nikita},
journal={arXiv preprint arXiv:2409.19300},
year={2024},
doi={10.2196/66919},
archivePrefix={arXiv},
eprint={2409.19300},
primaryClass={cs.SD cs.AI eess.AS}
}
|
ganitidis2024sustaining
|
arxiv-663079
|
2409.19301
|
Privacy Attack in Federated Learning is Not Easy: An Experimental Study
|
<|reference_start|>Privacy Attack in Federated Learning is Not Easy: An Experimental Study: Federated learning (FL) is an emerging distributed machine learning paradigm proposed for privacy preservation. Unlike traditional centralized learning approaches, FL enables multiple users to collaboratively train a shared global model without disclosing their own data, thereby significantly reducing the potential risk of privacy leakage. However, recent studies have indicated that FL cannot entirely guarantee privacy protection, and attackers may still be able to extract users' private data through the communicated model gradients. Although numerous privacy attack FL algorithms have been developed, most are designed to reconstruct private data from a single step of calculated gradients. It remains uncertain whether these methods are effective in realistic federated environments or if they have other limitations. In this paper, we aim to help researchers better understand and evaluate the effectiveness of privacy attacks on FL. We analyze and discuss recent research papers on this topic and conduct experiments in a real FL environment to compare the performance of various attack methods. Our experimental results reveal that none of the existing state-of-the-art privacy attack algorithms can effectively breach private client data in realistic FL settings, even in the absence of defense strategies. This suggests that privacy attacks in FL are more challenging than initially anticipated.<|reference_end|>
|
arxiv
|
@article{zhu2024privacy,
title={Privacy Attack in Federated Learning is Not Easy: An Experimental Study},
author={Hangyu Zhu, Liyuan Huang, Zhenping Xie},
journal={arXiv preprint arXiv:2409.19301},
year={2024},
archivePrefix={arXiv},
eprint={2409.19301},
primaryClass={cs.CR cs.AI}
}
|
zhu2024privacy
|
arxiv-663080
|
2409.19302
|
Leveraging MTD to Mitigate Poisoning Attacks in Decentralized FL with Non-IID Data
|
<|reference_start|>Leveraging MTD to Mitigate Poisoning Attacks in Decentralized FL with Non-IID Data: Decentralized Federated Learning (DFL), a paradigm for managing big data in a privacy-preserved manner, is still vulnerable to poisoning attacks where malicious clients tamper with data or models. Current defense methods often assume Independently and Identically Distributed (IID) data, which is unrealistic in real-world applications. In non-IID contexts, existing defensive strategies face challenges in distinguishing between models that have been compromised and those that have been trained on heterogeneous data distributions, leading to diminished efficacy. In response, this paper proposes a framework that employs the Moving Target Defense (MTD) approach to bolster the robustness of DFL models. By continuously modifying the attack surface of the DFL system, this framework aims to mitigate poisoning attacks effectively. The proposed MTD framework includes both proactive and reactive modes, utilizing a reputation system that combines metrics of model similarity and loss, alongside various defensive techniques. Comprehensive experimental evaluations indicate that the MTD-based mechanism significantly mitigates a range of poisoning attack types across multiple datasets with different topologies.<|reference_end|>
|
arxiv
|
@article{feng2024leveraging,
title={Leveraging MTD to Mitigate Poisoning Attacks in Decentralized FL with
Non-IID Data},
author={Chao Feng, Alberto Huertas Celdr'an, Zien Zeng, Zi Ye, Jan von der
Assen, Gerome Bovet, Burkhard Stiller},
journal={arXiv preprint arXiv:2409.19302},
year={2024},
archivePrefix={arXiv},
eprint={2409.19302},
primaryClass={cs.CR cs.DC}
}
|
feng2024leveraging
|
arxiv-663081
|
2409.19304
|
AVIATE: Exploiting Translation Variants of Artifacts to Improve IR-based Traceability Recovery in Bilingual Software Projects
|
<|reference_start|>AVIATE: Exploiting Translation Variants of Artifacts to Improve IR-based Traceability Recovery in Bilingual Software Projects: Traceability plays a vital role in facilitating various software development activities by establishing the traces between different types of artifacts (e.g., issues and commits in software repositories). Among the explorations for automated traceability recovery, the IR (Information Retrieval)-based approaches leverage textual similarity to measure the likelihood of traces between artifacts and show advantages in many scenarios. However, the globalization of software development has introduced new challenges, such as the possible multilingualism on the same concept (e.g., "ShuXing" vs. "attribute") in the artifact texts, thus significantly hampering the performance of IR-based approaches. Existing research has shown that machine translation can help address the term inconsistency in bilingual projects. However, the translation can also bring in synonymous terms that are not consistent with those in the bilingual projects (e.g., another translation of "ShuXing" as "property"). Therefore, we propose an enhancement strategy called AVIATE that exploits translation variants from different translators by utilizing the word pairs that appear simultaneously across the translation variants from different kinds artifacts (a.k.a. consensual biterms). We use these biterms to first enrich the artifact texts, and then to enhance the calculated IR values for improving IR-based traceability recovery for bilingual software projects. The experiments on 17 bilingual projects (involving English and 4 other languages) demonstrate that AVIATE significantly outperformed the IR-based approach with machine translation (the state-of-the-art in this field) with an average increase of 16.67 in Average Precision (31.43%) and 8.38 (11.22%) in Mean Average Precision, indicating its effectiveness in addressing the challenges of multilingual traceability recovery.<|reference_end|>
|
arxiv
|
@article{sun2024aviate:,
title={AVIATE: Exploiting Translation Variants of Artifacts to Improve IR-based
Traceability Recovery in Bilingual Software Projects},
author={Kexin Sun, Yiding Ren, Hongyu Kuang, Hui Gao, Xiaoxing Ma, Guoping
Rong, Dong Shao, He Zhang},
journal={arXiv preprint arXiv:2409.19304},
year={2024},
archivePrefix={arXiv},
eprint={2409.19304},
primaryClass={cs.SE}
}
|
sun2024aviate:
|
arxiv-663082
|
2409.19305
|
EEPNet: Efficient Edge Pixel-based Matching Network for Cross-Modal Dynamic Registration between LiDAR and Camera
|
<|reference_start|>EEPNet: Efficient Edge Pixel-based Matching Network for Cross-Modal Dynamic Registration between LiDAR and Camera: Multisensor fusion is essential for autonomous vehicles to accurately perceive, analyze, and plan their trajectories within complex environments. This typically involves the integration of data from LiDAR sensors and cameras, which necessitates high-precision and real-time registration. Current methods for registering LiDAR point clouds with images face significant challenges due to inherent modality differences and computational overhead. To address these issues, we propose EEPNet, an advanced network that leverages reflectance maps obtained from point cloud projections to enhance registration accuracy. The introduction of point cloud projections substantially mitigates cross-modality differences at the network input level, while the inclusion of reflectance data improves performance in scenarios with limited spatial information of point cloud within the camera's field of view. Furthermore, by employing edge pixels for feature matching and incorporating an efficient matching optimization layer, EEPNet markedly accelerates real-time registration tasks. Experimental validation demonstrates that EEPNet achieves superior accuracy and efficiency compared to state-of-the-art methods. Our contributions offer significant advancements in autonomous perception systems, paving the way for robust and efficient sensor fusion in real-world applications.<|reference_end|>
|
arxiv
|
@article{yue2024eepnet:,
title={EEPNet: Efficient Edge Pixel-based Matching Network for Cross-Modal
Dynamic Registration between LiDAR and Camera},
author={Yuanchao Yue, Hui Yuan, Suai Li, Qi Jiang},
journal={arXiv preprint arXiv:2409.19305},
year={2024},
archivePrefix={arXiv},
eprint={2409.19305},
primaryClass={cs.CV eess.IV}
}
|
yue2024eepnet:
|
arxiv-663083
|
2409.19306
|
CausalVE: Face Video Privacy Encryption via Causal Video Prediction
|
<|reference_start|>CausalVE: Face Video Privacy Encryption via Causal Video Prediction: Advanced facial recognition technologies and recommender systems with inadequate privacy technologies and policies for facial interactions increase concerns about bioprivacy violations. With the proliferation of video and live-streaming websites, public-face video distribution and interactions pose greater privacy risks. Existing techniques typically address the risk of sensitive biometric information leakage through various privacy enhancement methods but pose a higher security risk by corrupting the information to be conveyed by the interaction data, or by leaving certain biometric features intact that allow an attacker to infer sensitive biometric information from them. To address these shortcomings, in this paper, we propose a neural network framework, CausalVE. We obtain cover images by adopting a diffusion model to achieve face swapping with face guidance and use the speech sequence features and spatiotemporal sequence features of the secret video for dynamic video inference and prediction to obtain a cover video with the same number of frames as the secret video. In addition, we hide the secret video by using reversible neural networks for video hiding so that the video can also disseminate secret data. Numerous experiments prove that our CausalVE has good security in public video dissemination and outperforms state-of-the-art methods from a qualitative, quantitative, and visual point of view.<|reference_end|>
|
arxiv
|
@article{huang2024causalve:,
title={CausalVE: Face Video Privacy Encryption via Causal Video Prediction},
author={Yubo Huang, Wenhao Feng, Xin Lai, Zixi Wang, Jingzehua Xu, Shuai
Zhang, Hongjie He, Fan Chen},
journal={arXiv preprint arXiv:2409.19306},
year={2024},
archivePrefix={arXiv},
eprint={2409.19306},
primaryClass={cs.CV cs.AI}
}
|
huang2024causalve:
|
arxiv-663084
|
2409.19308
|
Designing Domain-Specific Large Language Models: The Critical Role of Fine-Tuning in Public Opinion Simulation
|
<|reference_start|>Designing Domain-Specific Large Language Models: The Critical Role of Fine-Tuning in Public Opinion Simulation: Large language models (LLMs) have transformed natural language processing across diverse fields, yet their general-purpose design limits their effectiveness in specialized domains, such as simulating opinions on environmental policies. This paper presents an approach for fine-tuning LLMs using data from the UK Household Longitudinal Study, improving the accuracy of opinion generation by conditioning models on socio-demographic factors like age, income, education, and region. By emulating diverse synthetic profiles, fine-tuned models capture the subtle differences across demographic groups more effectively than pre-trained versions. Metrics such as Chi-Squared, Cosine Similarity, Jaccard Index, and KL-divergence, demonstrate a strong alignment between synthetic and real-world opinion data. This approach highlights the potential of fine-tuning LLMs to provide more informed, representative, and ethical insights into public sentiments on environmental issues. The findings underscore the importance of tailoring LLMs to specific societal contexts for more accurate and ethical policy simulations.<|reference_end|>
|
arxiv
|
@article{lin2024designing,
title={Designing Domain-Specific Large Language Models: The Critical Role of
Fine-Tuning in Public Opinion Simulation},
author={Haocheng Lin},
journal={arXiv preprint arXiv:2409.19308},
year={2024},
archivePrefix={arXiv},
eprint={2409.19308},
primaryClass={cs.CL cs.AI}
}
|
lin2024designing
|
arxiv-663085
|
2409.19309
|
Temporal Consistency of Data and Information in Cyber-Physical Systems
|
<|reference_start|>Temporal Consistency of Data and Information in Cyber-Physical Systems: In a large cyber-physical system, a temporal inconsistency of an output value can arise if there is a non-negligible delay between the instant when a sensor value is acquired from the environment and the instant when a setpoint, based on this sensor value, is used in the environment. Such a temporal inconsistency can be the cause of a critical malfunction of the cyber-physical system. This paper presents a solution of this temporal consistency problem that can best be implemented in a time-triggered architecture (TTA). In a TTA, the instants of sensor value acquisition, setpoint calculation, and actuation on the environment are statically configured, and the cyber-physical system implements software and hardware mechanisms to execute the respective actions tightly at these configured instants.<|reference_end|>
|
arxiv
|
@article{kopetz2024temporal,
title={Temporal Consistency of Data and Information in Cyber-Physical Systems},
author={Hermann Kopetz (1) and Wilfried Steiner (2) ((1) TU Wien, Austria, (2)
TTTech, Austria)},
journal={arXiv preprint arXiv:2409.19309},
year={2024},
archivePrefix={arXiv},
eprint={2409.19309},
primaryClass={cs.NI}
}
|
kopetz2024temporal
|
arxiv-663086
|
2409.19310
|
Model X-Ray: Detection of Hidden Malware in AI Model Weights using Few Shot Learning
|
<|reference_start|>Model X-Ray: Detection of Hidden Malware in AI Model Weights using Few Shot Learning: The potential for exploitation of AI models has increased due to the rapid advancement of Artificial Intelligence (AI) and the widespread use of platforms like Model Zoo for sharing AI models. Attackers can embed malware within AI models through steganographic techniques, taking advantage of the substantial size of these models to conceal malicious data and use it for nefarious purposes, e.g. Remote Code Execution. Ensuring the security of AI models is a burgeoning area of research essential for safeguarding the multitude of organizations and users relying on AI technologies. This study leverages well-studied image few-shot learning techniques by transferring the AI models to the image field using a novel image representation. Applying few-shot learning in this field enables us to create practical models, a feat that previous works lack. Our method addresses critical limitations in state-of-the-art detection techniques that hinder their practicality. This approach reduces the required training dataset size from 40000 models to just 6. Furthermore, our methods consistently detect delicate attacks of up to 25% embedding rate and even up to 6% in some cases, while previous works were only shown to be effective for a 100%-50% embedding rate. We employ a strict evaluation strategy to ensure the trained models are generic concerning various factors. In addition, we show that our trained models successfully detect novel spread-spectrum steganography attacks, demonstrating the models' impressive robustness just by learning one type of attack. We open-source our code to support reproducibility and enhance the research in this new field.<|reference_end|>
|
arxiv
|
@article{gilkarov2024model,
title={Model X-Ray: Detection of Hidden Malware in AI Model Weights using Few
Shot Learning},
author={Daniel Gilkarov and Ran Dubin},
journal={arXiv preprint arXiv:2409.19310},
year={2024},
archivePrefix={arXiv},
eprint={2409.19310},
primaryClass={cs.CR cs.AI}
}
|
gilkarov2024model
|
arxiv-663087
|
2409.19315
|
Analog In-Memory Computing Attention Mechanism for Fast and Energy-Efficient Large Language Models
|
<|reference_start|>Analog In-Memory Computing Attention Mechanism for Fast and Energy-Efficient Large Language Models: Transformer neural networks, driven by self-attention mechanisms, are core components of foundational and Large Language Models. In generative transformers, self-attention uses cache memory to store token projections, avoiding recomputation at each time step. However, GPU-stored projections must be loaded into SRAM for each new generation step, causing latency and energy bottlenecks for long sequences. In this work, we propose a fast and energy-efficient hardware implementation of self-attention using analog in-memory computing based on gain cell memories. Volatile gain cell memories can be efficiently written to store new tokens during sequence generation, while performing analog signed weight multiplications to compute the dot-products required for self-attention. We implement Sliding Window Attention, which keeps memory of a finite set of past steps. A charge-to-pulse converter for array readout eliminates the need for analog-to-digital conversion between self-attention stages. Using a co-designed initialization algorithm to adapt pre-trained weights to gain cell non-idealities, we achieve NLP performance comparable to ChatGPT-2 with minimal training iterations, despite hardware constraints. Our end-to-end hardware design includes digital controls, estimating area, latency, and energy. The system reduces attention latency by up to two orders of magnitude and energy consumption by up to five orders compared to GPUs, marking a significant step toward ultra-fast, low-power sequence generation in Large Language Models.<|reference_end|>
|
arxiv
|
@article{leroux2024analog,
title={Analog In-Memory Computing Attention Mechanism for Fast and
Energy-Efficient Large Language Models},
author={Nathan Leroux, Paul-Philipp Manea, Chirag Sudarshan, Jan Finkbeiner,
Sebastian Siegel, John Paul Strachan, Emre Neftci},
journal={arXiv preprint arXiv:2409.19315},
year={2024},
archivePrefix={arXiv},
eprint={2409.19315},
primaryClass={cs.NE cs.AI cs.AR cs.ET}
}
|
leroux2024analog
|
arxiv-663088
|
2409.19316
|
Movable Antenna Enabled Near-Field Communications: Channel Modeling and Performance Optimization
|
<|reference_start|>Movable Antenna Enabled Near-Field Communications: Channel Modeling and Performance Optimization: Movable antenna (MA) technology offers promising potential to enhance wireless communication by allowing flexible antenna movement. To maximize spatial degrees of freedom (DoFs), larger movable regions are required, which may render the conventional far-field assumption for channels between transceivers invalid. In light of it, we investigate in this paper MA-enabled near-field communications, where a base station (BS) with multiple movable subarrays serves multiple users, each equipped with a fixed-position antenna (FPA). First, we extend the field response channel model for MA systems to the near-field propagation scenario. Next, we examine MA-aided multiuser communication systems under both digital and analog beamforming architectures. For digital beamforming, spatial division multiple access (SDMA) is utilized, where an upper bound on the minimum signal-to-interference-plus-noise ratio (SINR) across users is derived in closed form. A low-complexity algorithm based on zero-forcing (ZF) is then proposed to jointly optimize the antenna position vector (APV) and digital beamforming matrix (DBFM) to approach this bound. For analog beamforming, orthogonal frequency division multiple access (OFDMA) is employed, and an upper bound on the minimum signal-to-noise ratio (SNR) among users is derived. An alternating optimization (AO) algorithm is proposed to iteratively optimize the APV, analog beamforming vector (ABFV), and power allocation until convergence. For both architectures, we further explore MA design strategies based on statistical channel state information (CSI), with the APV updated less frequently to reduce the antenna movement overhead. Simulation results demonstrate that our proposed algorithms achieve performance close to the derived bounds and also outperform the benchmark schemes using dense or sparse arrays with FPAs.<|reference_end|>
|
arxiv
|
@article{zhu2024movable,
title={Movable Antenna Enabled Near-Field Communications: Channel Modeling and
Performance Optimization},
author={Lipeng Zhu, Wenyan Ma, Zhenyu Xiao, and Rui Zhang},
journal={arXiv preprint arXiv:2409.19316},
year={2024},
archivePrefix={arXiv},
eprint={2409.19316},
primaryClass={cs.IT eess.SP math.IT}
}
|
zhu2024movable
|
arxiv-663089
|
2409.19318
|
Fairness Analysis with Shapley-Owen Effects
|
<|reference_start|>Fairness Analysis with Shapley-Owen Effects: We argue that relative importance and its equitable attribution in terms of Shapley-Owen effects is an appropriate one, and, if we accept a small number of reasonable imperatives for equitable attribution, the only way to measure fairness. On the other hand, the computation of Shapley-Owen effects can be very demanding. Our main technical result is a spectral decomposition of the Shapley-Owen effects, which decomposes the computation of these indices into a model-specific and a model-independent part. The model-independent part is precomputed once and for all, and the model-specific computation of Shapley-Owen effects is expressed analytically in terms of the coefficients of the model's \emph{polynomial chaos expansion} (PCE), which can now be reused to compute different Shapley-Owen effects. We also propose an algorithm for computing precise and sparse truncations of the PCE of the model and the spectral decomposition of the Shapley-Owen effects, together with upper bounds on the accumulated approximation errors. The approximations of both the PCE and the Shapley-Owen effects converge to their true values.<|reference_end|>
|
arxiv
|
@article{ruess2024fairness,
title={Fairness Analysis with Shapley-Owen Effects},
author={Harald Ruess},
journal={arXiv preprint arXiv:2409.19318},
year={2024},
archivePrefix={arXiv},
eprint={2409.19318},
primaryClass={cs.AI cs.GT}
}
|
ruess2024fairness
|
arxiv-663090
|
2409.19322
|
Scalable Cloud-Native Pipeline for Efficient 3D Model Reconstruction from Monocular Smartphone Images
|
<|reference_start|>Scalable Cloud-Native Pipeline for Efficient 3D Model Reconstruction from Monocular Smartphone Images: In recent years, 3D models have gained popularity in various fields, including entertainment, manufacturing, and simulation. However, manually creating these models can be a time-consuming and resource-intensive process, making it impractical for large-scale industrial applications. To address this issue, researchers are exploiting Artificial Intelligence and Machine Learning algorithms to automatically generate 3D models effortlessly. In this paper, we present a novel cloud-native pipeline that can automatically reconstruct 3D models from monocular 2D images captured using a smartphone camera. Our goal is to provide an efficient and easily-adoptable solution that meets the Industry 4.0 standards for creating a Digital Twin model, which could enhance personnel expertise through accelerated training. We leverage machine learning models developed by NVIDIA Research Labs alongside a custom-designed pose recorder with a unique pose compensation component based on the ARCore framework by Google. Our solution produces a reusable 3D model, with embedded materials and textures, exportable and customizable in any external 3D modelling software or 3D engine. Furthermore, the whole workflow is implemented by adopting the microservices architecture standard, enabling each component of the pipeline to operate as a standalone replaceable module.<|reference_end|>
|
arxiv
|
@article{aghilar2024scalable,
title={Scalable Cloud-Native Pipeline for Efficient 3D Model Reconstruction
from Monocular Smartphone Images},
author={Potito Aghilar, Vito Walter Anelli, Michelantonio Trizio, Tommaso Di
Noia},
journal={arXiv preprint arXiv:2409.19322},
year={2024},
archivePrefix={arXiv},
eprint={2409.19322},
primaryClass={cs.CV cs.AI}
}
|
aghilar2024scalable
|
arxiv-663091
|
2409.19323
|
Intelligent Fish Detection System with Similarity-Aware Transformer
|
<|reference_start|>Intelligent Fish Detection System with Similarity-Aware Transformer: Fish detection in water-land transfer has significantly contributed to the fishery. However, manual fish detection in crowd-collaboration performs inefficiently and expensively, involving insufficient accuracy. To further enhance the water-land transfer efficiency, improve detection accuracy, and reduce labor costs, this work designs a new type of lightweight and plug-and-play edge intelligent vision system to automatically conduct fast fish detection with high-speed camera. Moreover, a novel similarity-aware vision Transformer for fast fish detection (FishViT) is proposed to onboard identify every single fish in a dense and similar group. Specifically, a novel similarity-aware multi-level encoder is developed to enhance multi-scale features in parallel, thereby yielding discriminative representations for varying-size fish. Additionally, a new soft-threshold attention mechanism is introduced, which not only effectively eliminates background noise from images but also accurately recognizes both the edge details and overall features of different similar fish. 85 challenging video sequences with high framerate and high-resolution are collected to establish a benchmark from real fish water-land transfer scenarios. Exhaustive evaluation conducted with this challenging benchmark has proved the robustness and effectiveness of FishViT with over 80 FPS. Real work scenario tests validate the practicality of the proposed method. The code and demo video are available at https://github.com/vision4robotics/FishViT.<|reference_end|>
|
arxiv
|
@article{li2024intelligent,
title={Intelligent Fish Detection System with Similarity-Aware Transformer},
author={Shengchen Li, Haobo Zuo, Changhong Fu, Zhiyong Wang, Zhiqiang Xu},
journal={arXiv preprint arXiv:2409.19323},
year={2024},
archivePrefix={arXiv},
eprint={2409.19323},
primaryClass={cs.RO}
}
|
li2024intelligent
|
arxiv-663092
|
2409.19325
|
A Generalized Model for Multidimensional Intransitivity
|
<|reference_start|>A Generalized Model for Multidimensional Intransitivity: Intransitivity is a critical issue in pairwise preference modeling. It refers to the intransitive pairwise preferences between a group of players or objects that potentially form a cyclic preference chain and has been long discussed in social choice theory in the context of the dominance relationship. However, such multifaceted intransitivity between players and the corresponding player representations in high dimensions is difficult to capture. In this paper, we propose a probabilistic model that jointly learns each player's d-dimensional representation (d>1) and a dataset-specific metric space that systematically captures the distance metric in Rd over the embedding space. Interestingly, by imposing additional constraints in the metric space, our proposed model degenerates to former models used in intransitive representation learning. Moreover, we present an extensive quantitative investigation of the vast existence of intransitive relationships between objects in various real-world benchmark datasets. To our knowledge, this investigation is the first of this type. The predictive performance of our proposed method on different real-world datasets, including social choice, election, and online game datasets, shows that our proposed method outperforms several competing methods in terms of prediction accuracy.<|reference_end|>
|
arxiv
|
@article{duan2024a,
title={A Generalized Model for Multidimensional Intransitivity},
author={Jiuding Duan, Jiyi Li, Yukino Baba, and Hisashi Kashima},
journal={arXiv preprint arXiv:2409.19325},
year={2024},
doi={10.1007/978-3-319-57529-2_65},
archivePrefix={arXiv},
eprint={2409.19325},
primaryClass={cs.LG cs.AI cs.CL cs.GT econ.GN q-fin.EC}
}
|
duan2024a
|
arxiv-663093
|
2409.19330
|
3D-CT-GPT: Generating 3D Radiology Reports through Integration of Large Vision-Language Models
|
<|reference_start|>3D-CT-GPT: Generating 3D Radiology Reports through Integration of Large Vision-Language Models: Medical image analysis is crucial in modern radiological diagnostics, especially given the exponential growth in medical imaging data. The demand for automated report generation systems has become increasingly urgent. While prior research has mainly focused on using machine learning and multimodal language models for 2D medical images, the generation of reports for 3D medical images has been less explored due to data scarcity and computational complexities. This paper introduces 3D-CT-GPT, a Visual Question Answering (VQA)-based medical visual language model specifically designed for generating radiology reports from 3D CT scans, particularly chest CTs. Extensive experiments on both public and private datasets demonstrate that 3D-CT-GPT significantly outperforms existing methods in terms of report accuracy and quality. Although current methods are few, including the partially open-source CT2Rep and the open-source M3D, we ensured fair comparison through appropriate data conversion and evaluation methodologies. Experimental results indicate that 3D-CT-GPT enhances diagnostic accuracy and report coherence, establishing itself as a robust solution for clinical radiology report generation. Future work will focus on expanding the dataset and further optimizing the model to enhance its performance and applicability.<|reference_end|>
|
arxiv
|
@article{chen20243d-ct-gpt:,
title={3D-CT-GPT: Generating 3D Radiology Reports through Integration of Large
Vision-Language Models},
author={Hao Chen, Wei Zhao, Yingli Li, Tianyang Zhong, Yisong Wang, Youlan
Shang, Lei Guo, Junwei Han, Tianming Liu, Jun Liu, Tuo Zhang},
journal={arXiv preprint arXiv:2409.19330},
year={2024},
archivePrefix={arXiv},
eprint={2409.19330},
primaryClass={cs.CV cs.AI}
}
|
chen20243d-ct-gpt:
|
arxiv-663094
|
2409.19334
|
OnePath: Efficient and Privacy-Preserving Decision Tree Inference in the Cloud
|
<|reference_start|>OnePath: Efficient and Privacy-Preserving Decision Tree Inference in the Cloud: The expansive storage capacity and robust computational power of cloud servers have led to the widespread outsourcing of machine learning inference services to the cloud. While this practice offers significant operational benefits, it also poses substantial privacy risks, including the exposure of proprietary models and sensitive user data. In this paper, we introduce OnePath, a framework designed for secure and efficient decision tree inference in cloud environments. Unlike existing schemes that require traversing all internal nodes of a decision tree, our protocol securely identifies and processes only the nodes on the prediction path, maintaining data privacy under ciphertext throughout the inference process. This selective traversal enhances both security and efficiency. To further optimize privacy and performance, OnePath employs lightweight cryptographic techniques, such as functional encryption, during the online phase of secure inference. Notably, our protocol allows both providers and clients to perform secure inference without the need to remain online continuously, a critical advantage for real-world applications. We substantiate the security of our framework with formal proofs, demonstrating that OnePath robustly protects the privacy of decision tree classifiers and user data. Experimental results highlight the efficiency of our approach, with our scheme processing query data in mere microseconds on the tested dataset. Through OnePath, we provide a practical solution that balances the needs for security and efficiency in cloud-based decision tree inference, making it a promising option for a variety of applications.<|reference_end|>
|
arxiv
|
@article{yuan2024onepath:,
title={OnePath: Efficient and Privacy-Preserving Decision Tree Inference in the
Cloud},
author={Shuai Yuan, Hongwei Li, Xinyuan Qian, Wenbo Jiang, Guowen Xu},
journal={arXiv preprint arXiv:2409.19334},
year={2024},
archivePrefix={arXiv},
eprint={2409.19334},
primaryClass={cs.CR}
}
|
yuan2024onepath:
|
arxiv-663095
|
2409.19337
|
Developing Cost-Effective Drones for 5G Non-Terrestrial Network Research and Experimentation
|
<|reference_start|>Developing Cost-Effective Drones for 5G Non-Terrestrial Network Research and Experimentation: In this article, we describe the components and procedures for building a drone ready for networking experimentation. In particular, our drone design includes multiple technologies and elements such as 4G/5G connectivity for real-time data transmission, a 360-degree camera for immersive vision and AR/VR, precise GPS for navigation, and a powerful Linux-based system with GPU for computer vision experiments and applications. Component selection and assembly techniques are included, along with software integration for a smooth, seamless operation of advanced edge applications.<|reference_end|>
|
arxiv
|
@article{cáceres2024developing,
title={Developing Cost-Effective Drones for 5G Non-Terrestrial Network Research
and Experimentation},
author={Carlos de Quinto C'aceres, Andr'es Navarro, Alejandro Leonardo
Garc'ia Navarro, Tom'as Mart'inez, Gabriel Otero, Jos'e Alberto
Hern'andez},
journal={arXiv preprint arXiv:2409.19337},
year={2024},
archivePrefix={arXiv},
eprint={2409.19337},
primaryClass={cs.AR}
}
|
cáceres2024developing
|
arxiv-663096
|
2409.19338
|
Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization in Social Networks
|
<|reference_start|>Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization in Social Networks: The impact of social media on critical issues such as echo chambers needs to be addressed, as these phenomena can have disruptive consequences for our society. Traditional research often oversimplifies emotional tendencies and opinion evolution into numbers and formulas, neglecting that news and communication are conveyed through text, which limits these approaches. Hence, in this work, we propose an LLM-based simulation for the social opinion network to evaluate and counter polarization phenomena. We first construct three typical network structures to simulate different characteristics of social interactions. Then, agents interact based on recommendation algorithms and update their strategies through reasoning and analysis. By comparing these interactions with the classic Bounded Confidence Model (BCM), the Friedkin Johnsen (FJ) model, and using echo chamber-related indices, we demonstrate the effectiveness of our framework in simulating opinion dynamics and reproducing phenomena such as opinion polarization and echo chambers. We propose two mitigation methods, active and passive nudges, that can help reduce echo chambers, specifically within language-based simulations. We hope our work will offer valuable insights and guidance for social polarization mitigation.<|reference_end|>
|
arxiv
|
@article{wang2024decoding,
title={Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization
in Social Networks},
author={Chenxi Wang, Zongfang Liu, Dequan Yang, Xiuying Chen},
journal={arXiv preprint arXiv:2409.19338},
year={2024},
archivePrefix={arXiv},
eprint={2409.19338},
primaryClass={cs.SI cs.CL}
}
|
wang2024decoding
|
arxiv-663097
|
2409.19339
|
Visual Question Decomposition on Multimodal Large Language Models
|
<|reference_start|>Visual Question Decomposition on Multimodal Large Language Models: Question decomposition has emerged as an effective strategy for prompting Large Language Models (LLMs) to answer complex questions. However, while existing methods primarily focus on unimodal language models, the question decomposition capability of Multimodal Large Language Models (MLLMs) has yet to be explored. To this end, this paper explores visual question decomposition on MLLMs. Specifically, we introduce a systematic evaluation framework including a dataset and several evaluation criteria to assess the quality of the decomposed sub-questions, revealing that existing MLLMs struggle to produce high-quality sub-questions. To address this limitation, we propose a specific finetuning dataset, DecoVQA+, for enhancing the model's question decomposition capability. Aiming at enabling models to perform appropriate selective decomposition, we propose an efficient finetuning pipeline. The finetuning pipeline consists of our proposed dataset and a training objective for selective decomposition. Finetuned MLLMs demonstrate significant improvements in the quality of sub-questions and the policy of selective question decomposition. Additionally, the models also achieve higher accuracy with selective decomposition on VQA benchmark datasets.<|reference_end|>
|
arxiv
|
@article{zhang2024visual,
title={Visual Question Decomposition on Multimodal Large Language Models},
author={Haowei Zhang, Jianzhe Liu, Zhen Han, Shuo Chen, Bailan He, Volker
Tresp, Zhiqiang Xu, Jindong Gu},
journal={arXiv preprint arXiv:2409.19339},
year={2024},
archivePrefix={arXiv},
eprint={2409.19339},
primaryClass={cs.CL cs.AI cs.CV cs.LG}
}
|
zhang2024visual
|
arxiv-663098
|
2409.19341
|
Analysis of the SiMPL method for density-based topology optimization
|
<|reference_start|>Analysis of the SiMPL method for density-based topology optimization: We present a rigorous convergence analysis of a new method for density-based topology optimization: Sigmoidal Mirror descent with a Projected Latent variable. SiMPL provides point-wise bound preserving design updates and faster convergence than other popular first-order topology optimization methods. Due to its strong bound preservation, the method is exceptionally robust, as demonstrated in numerous examples here and in a companion article. Furthermore, it is easy to implement with clear structure and analytical expressions for the updates. Our analysis covers two versions of the method, characterized by the employed line search strategies. We consider a modified Armijo backtracking line search and a Bregman backtracking line search. Regardless of the line search algorithm, SiMPL delivers a strict monotone decrease in the objective function and further intuitive convergence properties, e.g., strong and pointwise convergence of the density variables on the active sets, norm convergence to zero of the increments, and more. In addition, the numerical experiments demonstrate apparent mesh-independent convergence of the algorithm and superior performance over the two most popular first-order methods in topology optimization: OC and MMA.<|reference_end|>
|
arxiv
|
@article{keith2024analysis,
title={Analysis of the SiMPL method for density-based topology optimization},
author={Brendan Keith, Dohyun Kim, Boyan S. Lazarov, Thomas M. Surowiec},
journal={arXiv preprint arXiv:2409.19341},
year={2024},
archivePrefix={arXiv},
eprint={2409.19341},
primaryClass={math.OC cs.NA math.NA}
}
|
keith2024analysis
|
arxiv-663099
|
2409.19342
|
X-Prompt: Multi-modal Visual Prompt for Video Object Segmentation
|
<|reference_start|>X-Prompt: Multi-modal Visual Prompt for Video Object Segmentation: Multi-modal Video Object Segmentation (VOS), including RGB-Thermal, RGB-Depth, and RGB-Event, has garnered attention due to its capability to address challenging scenarios where traditional VOS methods struggle, such as extreme illumination, rapid motion, and background distraction. Existing approaches often involve designing specific additional branches and performing full-parameter fine-tuning for fusion in each task. However, this paradigm not only duplicates research efforts and hardware costs but also risks model collapse with the limited multi-modal annotated data. In this paper, we propose a universal framework named X-Prompt for all multi-modal video object segmentation tasks, designated as RGB+X. The X-Prompt framework first pre-trains a video object segmentation foundation model using RGB data, and then utilize the additional modality of the prompt to adapt it to downstream multi-modal tasks with limited data. Within the X-Prompt framework, we introduce the Multi-modal Visual Prompter (MVP), which allows prompting foundation model with the various modalities to segment objects precisely. We further propose the Multi-modal Adaptation Experts (MAEs) to adapt the foundation model with pluggable modality-specific knowledge without compromising the generalization capacity. To evaluate the effectiveness of the X-Prompt framework, we conduct extensive experiments on 3 tasks across 4 benchmarks. The proposed universal X-Prompt framework consistently outperforms the full fine-tuning paradigm and achieves state-of-the-art performance. Code: https://github.com/PinxueGuo/X-Prompt.git<|reference_end|>
|
arxiv
|
@article{guo2024x-prompt:,
title={X-Prompt: Multi-modal Visual Prompt for Video Object Segmentation},
author={Pinxue Guo, Wanyun Li, Hao Huang, Lingyi Hong, Xinyu Zhou, Zhaoyu
Chen, Jinglun Li, Kaixun Jiang, Wei Zhang, Wenqiang Zhang},
journal={arXiv preprint arXiv:2409.19342},
year={2024},
archivePrefix={arXiv},
eprint={2409.19342},
primaryClass={cs.CV}
}
|
guo2024x-prompt:
|
arxiv-663100
|
2409.19345
|
Unveil Benign Overfitting for Transformer in Vision: Training Dynamics, Convergence, and Generalization
|
<|reference_start|>Unveil Benign Overfitting for Transformer in Vision: Training Dynamics, Convergence, and Generalization: Transformers have demonstrated great power in the recent development of large foundational models. In particular, the Vision Transformer (ViT) has brought revolutionary changes to the field of vision, achieving significant accomplishments on the experimental side. However, their theoretical capabilities, particularly in terms of generalization when trained to overfit training data, are still not fully understood. To address this gap, this work delves deeply into the benign overfitting perspective of transformers in vision. To this end, we study the optimization of a Transformer composed of a self-attention layer with softmax followed by a fully connected layer under gradient descent on a certain data distribution model. By developing techniques that address the challenges posed by softmax and the interdependent nature of multiple weights in transformer optimization, we successfully characterized the training dynamics and achieved generalization in post-training. Our results establish a sharp condition that can distinguish between the small test error phase and the large test error regime, based on the signal-to-noise ratio in the data model. The theoretical results are further verified by experimental simulation.<|reference_end|>
|
arxiv
|
@article{jiang2024unveil,
title={Unveil Benign Overfitting for Transformer in Vision: Training Dynamics,
Convergence, and Generalization},
author={Jiarui Jiang, Wei Huang, Miao Zhang, Taiji Suzuki, Liqiang Nie},
journal={arXiv preprint arXiv:2409.19345},
year={2024},
archivePrefix={arXiv},
eprint={2409.19345},
primaryClass={cs.LG cs.CV stat.ML}
}
|
jiang2024unveil
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.