publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-10-05T17:36:16Z | http://arxiv.org/pdf/2310.03710v1 | http://arxiv.org/abs/2310.03710v1 | Nicholas Crispino, Kyle Montgomery, Fankun Zeng, Dawn Song, Chenguang Wang | 2023-10-05T17:36:16Z | cs.CL, cs.AI, cs.LG | We introduce a method to improve the zero-shot reasoning abilities of large
language models on general language understanding tasks. Specifically, we build
an autonomous agent to instruct the reasoning process of large language models.
We show this approach further unleashes the zero-shot reasoning abilities of
large language models to more tasks. We study the performance of our method on
a wide set of datasets spanning generation, classification, and reasoning. We
show that our method generalizes to most tasks and obtains state-of-the-art
zero-shot performance on 20 of the 29 datasets that we evaluate. For instance,
our method boosts the performance of state-of-the-art large language models by
a large margin, including Vicuna-13b (13.3%), Llama-2-70b-chat (23.2%), and
GPT-3.5 Turbo (17.0%). Compared to zero-shot chain of thought, our improvement
in reasoning is striking, with an average increase of 10.5%. With our method,
Llama-2-70b-chat outperforms zero-shot GPT-3.5 Turbo by 10.2%. | cs.CL | Agent Instructs Large Language Models to be General Zero-Shot Reasoners |
2023-10-05T17:12:17Z | http://arxiv.org/pdf/2310.03693v1 | http://arxiv.org/abs/2310.03693v1 | Xiangyu Qi, Yi Zeng, Tinghao Xie, Pin-Yu Chen, Ruoxi Jia, Prateek Mittal, Peter Henderson | 2023-10-05T17:12:17Z | cs.CL, cs.AI, cs.CR, cs.LG | Optimizing large language models (LLMs) for downstream use cases often
involves the customization of pre-trained LLMs through further fine-tuning.
Meta's open release of Llama models and OpenAI's APIs for fine-tuning GPT-3.5
Turbo on custom datasets also encourage this practice. But, what are the safety
costs associated with such custom fine-tuning? We note that while existing
safety alignment infrastructures can restrict harmful behaviors of LLMs at
inference time, they do not cover safety risks when fine-tuning privileges are
extended to end-users. Our red teaming studies find that the safety alignment
of LLMs can be compromised by fine-tuning with only a few adversarially
designed training examples. For instance, we jailbreak GPT-3.5 Turbo's safety
guardrails by fine-tuning it on only 10 such examples at a cost of less than
$0.20 via OpenAI's APIs, making the model responsive to nearly any harmful
instructions. Disconcertingly, our research also reveals that, even without
malicious intent, simply fine-tuning with benign and commonly used datasets can
also inadvertently degrade the safety alignment of LLMs, though to a lesser
extent. These findings suggest that fine-tuning aligned LLMs introduces new
safety risks that current safety infrastructures fall short of addressing --
even if a model's initial safety alignment is impeccable, it is not necessarily
to be maintained after custom fine-tuning. We outline and critically analyze
potential mitigations and advocate for further research efforts toward
reinforcing safety protocols for the custom fine-tuning of aligned LLMs. | cs.CL | Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! |
2023-10-05T17:04:59Z | http://arxiv.org/pdf/2310.03686v2 | http://arxiv.org/abs/2310.03686v2 | Anna Langedijk, Hosein Mohebbi, Gabriele Sarti, Willem Zuidema, Jaap Jumelet | 2024-04-03T12:09:26Z | cs.CL | In recent years, many interpretability methods have been proposed to help
interpret the internal states of Transformer-models, at different levels of
precision and complexity. Here, to analyze encoder-decoder Transformers, we
propose a simple, new method: DecoderLens. Inspired by the LogitLens (for
decoder-only Transformers), this method involves allowing the decoder to
cross-attend representations of intermediate encoder layers instead of using
the final encoder output, as is normally done in encoder-decoder models. The
method thus maps previously uninterpretable vector representations to
human-interpretable sequences of words or symbols. We report results from the
DecoderLens applied to models trained on question answering, logical reasoning,
speech recognition and machine translation. The DecoderLens reveals several
specific subtasks that are solved at low or intermediate layers, shedding new
light on the information flow inside the encoder component of this important
class of models. | cs.CL | DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers |
2023-10-05T16:43:13Z | http://arxiv.org/pdf/2310.03668v5 | http://arxiv.org/abs/2310.03668v5 | Oscar Sainz, Iker García-Ferrero, Rodrigo Agerri, Oier Lopez de Lacalle, German Rigau, Eneko Agirre | 2024-03-06T16:38:03Z | cs.CL | Large Language Models (LLMs) combined with instruction tuning have made
significant progress when generalizing to unseen tasks. However, they have been
less successful in Information Extraction (IE), lagging behind task-specific
models. Typically, IE tasks are characterized by complex annotation guidelines
that describe the task and give examples to humans. Previous attempts to
leverage such information have failed, even with the largest models, as they
are not able to follow the guidelines out of the box. In this paper, we propose
GoLLIE (Guideline-following Large Language Model for IE), a model able to
improve zero-shot results on unseen IE tasks by virtue of being fine-tuned to
comply with annotation guidelines. Comprehensive evaluation empirically
demonstrates that GoLLIE is able to generalize to and follow unseen guidelines,
outperforming previous attempts at zero-shot information extraction. The
ablation study shows that detailed guidelines are key for good results. | cs.CL | GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction |
2023-10-05T16:43:04Z | http://arxiv.org/pdf/2310.03666v1 | http://arxiv.org/abs/2310.03666v1 | Nicolas Matentzoglu, J. Harry Caufield, Harshad B. Hegde, Justin T. Reese, Sierra Moxon, Hyeongsik Kim, Nomi L. Harris, Melissa A Haendel, Christopher J. Mungall | 2023-10-05T16:43:04Z | cs.CL, cs.AI | Aligning terminological resources, including ontologies, controlled
vocabularies, taxonomies, and value sets is a critical part of data integration
in many domains such as healthcare, chemistry, and biomedical research. Entity
mapping is the process of determining correspondences between entities across
these resources, such as gene identifiers, disease concepts, or chemical entity
identifiers. Many tools have been developed to compute such mappings based on
common structural features and lexical information such as labels and synonyms.
Lexical approaches in particular often provide very high recall, but low
precision, due to lexical ambiguity. As a consequence of this, mapping efforts
often resort to a labor intensive manual mapping refinement through a human
curator.
Large Language Models (LLMs), such as the ones employed by ChatGPT, have
generalizable abilities to perform a wide range of tasks, including
question-answering and information extraction. Here we present MapperGPT, an
approach that uses LLMs to review and refine mapping relationships as a
post-processing step, in concert with existing high-recall methods that are
based on lexical and structural heuristics.
We evaluated MapperGPT on a series of alignment tasks from different domains,
including anatomy, developmental biology, and renal diseases. We devised a
collection of tasks that are designed to be particularly challenging for
lexical methods. We show that when used in combination with high-recall
methods, MapperGPT can provide a substantial improvement in accuracy, beating
state-of-the-art (SOTA) methods such as LogMap. | cs.CL | MapperGPT: Large Language Models for Linking and Mapping Entities |
2023-10-05T16:21:36Z | http://arxiv.org/pdf/2310.03646v2 | http://arxiv.org/abs/2310.03646v2 | Tom Sherborne, Naomi Saphra, Pradeep Dasigi, Hao Peng | 2024-03-12T13:38:31Z | cs.LG, cs.CL | Sharpness-aware minimization (SAM) reports improving domain generalization by
reducing the loss surface curvature in the parameter space. However,
generalization during fine-tuning is often more dependent on the
transferability of representations in the function space. Trust-region methods
(TR) target this goal by regularizing representation curvature to reduce
catastrophic forgetting of pre-trained task-agnostic information while adopting
task-specific skills. We consider unifying these strategies for low curvature
in both parameter space and function space to improve out-of-domain (OOD)
generalization. We propose Trust Region Aware Minimization (TRAM), a SAM
algorithm fine-tuning for low parameter sharpness and smooth, informative
representations preserving pre-trained structure. TRAM uses a trust region
bound to inform the SAM adversarial neighborhood, introducing an awareness of
function curvature within optimization for flatter minima. We empirically
validate TRAM in vision (cross-dataset adaptation) and text (OOD language
modeling, zero-shot cross-lingual transfer) tasks where robust domain transfer
and representation generality are critical. TRAM outperforms SAM- and TR-based
optimization across all tasks, notably surpassing competing methods for hard
transfer between anticorrelated domains. TRAM establishes a novel standard in
fine-tuning for domain-generalizable models with minimal additional computation
over previous sharpness-aware methods. | cs.LG | TRAM: Bridging Trust Regions and Sharpness Aware Minimization |
2023-10-05T16:14:46Z | http://arxiv.org/pdf/2310.03779v1 | http://arxiv.org/abs/2310.03779v1 | Yanming Wan, Jiayuan Mao, Joshua B. Tenenbaum | 2023-10-05T16:14:46Z | cs.AI, cs.CL, cs.LG, cs.RO | We introduce HandMeThat, a benchmark for a holistic evaluation of instruction
understanding and following in physical and social environments. While previous
datasets primarily focused on language grounding and planning, HandMeThat
considers the resolution of human instructions with ambiguities based on the
physical (object states and relations) and social (human actions and goals)
information. HandMeThat contains 10,000 episodes of human-robot interactions.
In each episode, the robot first observes a trajectory of human actions towards
her internal goal. Next, the robot receives a human instruction and should take
actions to accomplish the subgoal set through the instruction. In this paper,
we present a textual interface for our benchmark, where the robot interacts
with a virtual environment through textual commands. We evaluate several
baseline models on HandMeThat, and show that both offline and online
reinforcement learning algorithms perform poorly on HandMeThat, suggesting
significant room for future work on physical and social human-robot
communications and interactions. | cs.AI | HandMeThat: Human-Robot Communication in Physical and Social Environments |
2023-10-05T16:11:14Z | http://arxiv.org/pdf/2310.03639v2 | http://arxiv.org/abs/2310.03639v2 | Chih-Chen Chen, William Chen, Rodolfo Zevallos, John E. Ortega | 2023-10-08T23:28:50Z | cs.CL, eess.AS | The application of self-supervision to speech representation learning has
garnered significant interest in recent years, due to its scalability to large
amounts of unlabeled data. However, much progress, both in terms of
pre-training and downstream evaluation, has remained concentrated in
monolingual models that only consider English. Few models consider other
languages, and even fewer consider indigenous ones. In our submission to the
New Language Track of the ASRU 2023 ML-SUPERB Challenge, we present an ASR
corpus for Quechua, an indigenous South American Language. We benchmark the
efficacy of large SSL models on Quechua, along with 6 other indigenous
languages such as Guarani and Bribri, on low-resource ASR. Our results show
surprisingly strong performance by state-of-the-art SSL models, showing the
potential generalizability of large-scale models to real-world data. | cs.CL | Evaluating Self-Supervised Speech Representations for Indigenous American Languages |
2023-10-05T16:09:48Z | http://arxiv.org/pdf/2310.03635v1 | http://arxiv.org/abs/2310.03635v1 | Jiayuan Mao, Xuelin Yang, Xikun Zhang, Noah D. Goodman, Jiajun Wu | 2023-10-05T16:09:48Z | cs.AI, cs.CL, cs.CV, cs.LG, stat.ML | Building machines that can reason about physical events and their causal
relationships is crucial for flexible interaction with the physical world.
However, most existing physical and causal reasoning benchmarks are exclusively
based on synthetically generated events and synthetic natural language
descriptions of causal relationships. This design brings up two issues. First,
there is a lack of diversity in both event types and natural language
descriptions; second, causal relationships based on manually-defined heuristics
are different from human judgments. To address both shortcomings, we present
the CLEVRER-Humans benchmark, a video reasoning dataset for causal judgment of
physical events with human labels. We employ two techniques to improve data
collection efficiency: first, a novel iterative event cloze task to elicit a
new representation of events in videos, which we term Causal Event Graphs
(CEGs); second, a data augmentation technique based on neural language
generative models. We convert the collected CEGs into questions and answers to
be consistent with prior work. Finally, we study a collection of baseline
approaches for CLEVRER-Humans question-answering, highlighting the great
challenges set forth by our benchmark. | cs.AI | CLEVRER-Humans: Describing Physical and Causal Events the Human Way |
2023-10-05T15:23:04Z | http://arxiv.org/pdf/2310.04465v1 | http://arxiv.org/abs/2310.04465v1 | Helena Mihaljević, Elisabeth Steffen | 2023-10-05T15:23:04Z | cs.CL, cs.AI, cs.CY | The Perspective API, a popular text toxicity assessment service by Google and
Jigsaw, has found wide adoption in several application areas, notably content
moderation, monitoring, and social media research. We examine its potentials
and limitations for the detection of antisemitic online content that, by
definition, falls under the toxicity umbrella term. Using a manually annotated
German-language dataset comprising around 3,600 posts from Telegram and
Twitter, we explore as how toxic antisemitic texts are rated and how the
toxicity scores differ regarding different subforms of antisemitism and the
stance expressed in the texts. We show that, on a basic level, Perspective API
recognizes antisemitic content as toxic, but shows critical weaknesses with
respect to non-explicit forms of antisemitism and texts taking a critical
stance towards it. Furthermore, using simple text manipulations, we demonstrate
that the use of widespread antisemitic codes can substantially reduce API
scores, making it rather easy to bypass content moderation based on the
service's results. | cs.CL | How toxic is antisemitism? Potentials and limitations of automated toxicity scoring for antisemitic online content |
2023-10-05T14:18:40Z | http://arxiv.org/pdf/2310.03560v3 | http://arxiv.org/abs/2310.03560v3 | Fergus Imrie, Paulius Rauba, Mihaela van der Schaar | 2024-02-29T18:37:40Z | cs.CL | Digital health tools have the potential to significantly improve the delivery
of healthcare services. However, their adoption remains comparatively limited
due, in part, to challenges surrounding usability and trust. Large Language
Models (LLMs) have emerged as general-purpose models with the ability to
process complex information and produce human-quality text, presenting a wealth
of potential applications in healthcare. Directly applying LLMs in clinical
settings is not straightforward, however, with LLMs susceptible to providing
inconsistent or nonsensical answers. We demonstrate how LLM-based systems can
utilize external tools and provide a novel interface between clinicians and
digital technologies. This enhances the utility and practical impact of digital
healthcare tools and AI models while addressing current issues with using LLMs
in clinical settings such as hallucinations. We illustrate LLM-based interfaces
with the example of cardiovascular disease risk prediction. We develop a new
prognostic tool using automated machine learning and demonstrate how LLMs can
provide a unique interface to both our model and existing risk scores,
highlighting the benefit compared to traditional interfaces for digital tools. | cs.CL | Redefining Digital Health Interfaces with Large Language Models |
2023-10-05T12:59:57Z | http://arxiv.org/pdf/2310.03518v1 | http://arxiv.org/abs/2310.03518v1 | Jiachi Liu, Liwen Wang, Guanting Dong, Xiaoshuai Song, Zechen Wang, Zhengyang Wang, Shanglin Lei, Jinzheng Zhao, Keqing He, Bo Xiao, Weiran Xu | 2023-10-05T12:59:57Z | cs.CL, cs.AI, cs.DS | In real dialogue scenarios, as there are unknown input noises in the
utterances, existing supervised slot filling models often perform poorly in
practical applications. Even though there are some studies on noise-robust
models, these works are only evaluated on rule-based synthetic datasets, which
is limiting, making it difficult to promote the research of noise-robust
methods. In this paper, we introduce a noise robustness evaluation dataset
named Noise-SF for slot filling task. The proposed dataset contains five types
of human-annotated noise, and all those noises are exactly existed in real
extensive robust-training methods of slot filling into the proposed framework.
By conducting exhaustive empirical evaluation experiments on Noise-SF, we find
that baseline models have poor performance in robustness evaluation, and the
proposed framework can effectively improve the robustness of models. Based on
the empirical experimental results, we make some forward-looking suggestions to
fuel the research in this direction. Our dataset Noise-SF will be released at
https://github.com/dongguanting/Noise-SF. | cs.CL | Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations |
2023-10-05T12:13:00Z | http://arxiv.org/pdf/2310.03777v1 | http://arxiv.org/abs/2310.03777v1 | Saifullah Saifullah, Stefan Agne, Andreas Dengel, Sheraz Ahmed | 2023-10-05T12:13:00Z | cs.CL | In this paper, we introduce strategies for developing private Key Information
Extraction (KIE) systems by leveraging large pretrained document foundation
models in conjunction with differential privacy (DP), federated learning (FL),
and Differentially Private Federated Learning (DP-FL). Through extensive
experimentation on six benchmark datasets (FUNSD, CORD, SROIE, WildReceipts,
XFUND, and DOCILE), we demonstrate that large document foundation models can be
effectively fine-tuned for the KIE task under private settings to achieve
adequate performance while maintaining strong privacy guarantees. Moreover, by
thoroughly analyzing the impact of various training and model parameters on
model performance, we propose simple yet effective guidelines for achieving an
optimal privacy-utility trade-off for the KIE task under global DP. Finally, we
introduce FeAm-DP, a novel DP-FL algorithm that enables efficiently upscaling
global DP from a standalone context to a multi-client federated environment. We
conduct a comprehensive evaluation of the algorithm across various client and
privacy settings, and demonstrate its capability to achieve comparable
performance and privacy guarantees to standalone DP, even when accommodating an
increasing number of participating clients. Overall, our study offers valuable
insights into the development of private KIE systems, and highlights the
potential of document foundation models for privacy-preserved Document AI
applications. To the best of authors' knowledge, this is the first work that
explores privacy preserved document KIE using document foundation models. | cs.CL | PrIeD-KIE: Towards Privacy Preserved Document Key Information Extraction |
2023-10-05T11:45:29Z | http://arxiv.org/pdf/2310.03477v1 | http://arxiv.org/abs/2310.03477v1 | François Remy, Pieter Delobelle, Bettina Berendt, Kris Demuynck, Thomas Demeester | 2023-10-05T11:45:29Z | cs.CL, cs.AI | Training monolingual language models for low and mid-resource languages is
made challenging by limited and often inadequate pretraining data. In this
study, we propose a novel model conversion strategy to address this issue,
adapting high-resources monolingual language models to a new target language.
By generalizing over a word translation dictionary encompassing both the source
and target languages, we map tokens from the target tokenizer to semantically
similar tokens from the source language tokenizer. This one-to-many token
mapping improves tremendously the initialization of the embedding table for the
target language. We conduct experiments to convert high-resource models to mid-
and low-resource languages, namely Dutch and Frisian. These converted models
achieve a new state-of-the-art performance on these languages across all sorts
of downstream tasks. By reducing significantly the amount of data and time
required for training state-of-the-art models, our novel model conversion
strategy has the potential to benefit many languages worldwide. | cs.CL | Tik-to-Tok: Translating Language Models One Token at a Time: An Embedding Initialization Strategy for Efficient Language Adaptation |
2023-10-05T11:29:09Z | http://arxiv.org/pdf/2310.03473v1 | http://arxiv.org/abs/2310.03473v1 | Litton J Kurisinkel, Nancy F chen | 2023-10-05T11:29:09Z | cs.CL | Memory-efficient large language models are good at refining text input for
better readability. However, controllability is a matter of concern when it
comes to text generation tasks with long inputs, such as multi-document
summarization. In this work, we investigate for a generic controllable approach
for multi-document summarization that leverages the capabilities of LLMs to
refine the text. In particular, we train a controllable content extraction
scheme to extract the text that will be refined by an LLM. The scheme is
designed with a novel coverage and coherence intuitive policy, which is duly
rewarded by a passively trained LLM. Our approach yields competitive results in
the evaluation using ROUGE metrics and outperforms potential baselines in
coherence, as per human evaluation. | cs.CL | Controllable Multi-document Summarization: Coverage & Coherence Intuitive Policy with Large Language Model Based Rewards |
2023-10-05T10:29:18Z | http://arxiv.org/pdf/2310.03443v2 | http://arxiv.org/abs/2310.03443v2 | Li-Wei Chen, Kai-Chen Cheng, Hung-Shin Lee | 2023-10-06T02:54:17Z | cs.CL, cs.SD, eess.AS | This report provides a concise overview of the proposed North system, which
aims to achieve automatic word/syllable recognition for Taiwanese Hakka
(Sixian). The report outlines three key components of the system: the
acquisition, composition, and utilization of the training data; the
architecture of the model; and the hardware specifications and operational
statistics. The demonstration of the system has been made public at
https://asrvm.iis.sinica.edu.tw/hakka_sixian. | cs.CL | The North System for Formosa Speech Recognition Challenge 2023 |
2023-10-05T10:01:32Z | http://arxiv.org/pdf/2310.03424v1 | http://arxiv.org/abs/2310.03424v1 | Leonardo Emili, Thiago Fraga-Silva, Ernest Pusateri, Markus Nußbaum-Thom, Youssef Oualil | 2023-10-05T10:01:32Z | cs.LG, cs.CL | We study model pruning methods applied to Transformer-based neural network
language models for automatic speech recognition. We explore three aspects of
the pruning frame work, namely criterion, method and scheduler, analyzing their
contribution in terms of accuracy and inference speed. To the best of our
knowledge, such in-depth analyses on large-scale recognition systems has not
been reported in the literature. In addition, we propose a variant of low-rank
approximation suitable for incrementally compressing models, and delivering
multiple models with varied target sizes. Among other results, we show that a)
data-driven pruning outperforms magnitude-driven in several scenarios; b)
incremental pruning achieves higher accuracy compared to one-shot pruning,
especially when targeting smaller sizes; and c) low-rank approximation presents
the best trade-off between size reduction and inference speed-up for moderate
compression. | cs.LG | Neural Language Model Pruning for Automatic Speech Recognition |
2023-10-05T09:38:09Z | http://arxiv.org/pdf/2310.03414v1 | http://arxiv.org/abs/2310.03414v1 | Litton J Kurisinkel, Nancy F. Chen | 2023-10-05T09:38:09Z | cs.CL | Multi-document summarization is a challenging task due to its inherent
subjective bias, highlighted by the low inter-annotator ROUGE-1 score of 0.4
among DUC-2004 reference summaries. In this work, we aim to enhance the
objectivity of news summarization by focusing on the main event of a group of
related news documents and presenting it coherently with sufficient context.
Our primary objective is to succinctly report the main event, ensuring that the
summary remains objective and informative. To achieve this, we employ an
extract-rewrite approach that incorporates a main-event biased
monotone-submodular function for content selection. This enables us to extract
the most crucial information related to the main event from the document
cluster. To ensure coherence, we utilize a fine-tuned Language Model (LLM) for
rewriting the extracted content into a coherent text. The evaluation using
objective metrics and human evaluators confirms the effectiveness of our
approach, as it surpasses potential baselines, demonstrating excellence in both
content coverage, coherence, and informativeness. | cs.CL | LLM Based Multi-Document Summarization Exploiting Main-Event Biased Monotone Submodular Content Extraction |
2023-10-05T08:27:33Z | http://arxiv.org/pdf/2310.03376v1 | http://arxiv.org/abs/2310.03376v1 | Anisa Rula, Jennifer D'Souza | 2023-10-05T08:27:33Z | cs.CL, cs.AI, cs.IT, math.IT | Recent advancements in the field of Natural Language Processing, particularly
the development of large-scale language models that are pretrained on vast
amounts of knowledge, are creating novel opportunities within the realm of
Knowledge Engineering. In this paper, we investigate the usage of large
language models (LLMs) in both zero-shot and in-context learning settings to
tackle the problem of extracting procedures from unstructured PDF text in an
incremental question-answering fashion. In particular, we leverage the current
state-of-the-art GPT-4 (Generative Pre-trained Transformer 4) model,
accompanied by two variations of in-context learning that involve an ontology
with definitions of procedures and steps and a limited number of samples of
few-shot learning. The findings highlight both the promise of this approach and
the value of the in-context learning customisations. These modifications have
the potential to significantly address the challenge of obtaining sufficient
training data, a hurdle often encountered in deep learning-based Natural
Language Processing techniques for procedure extraction. | cs.CL | Procedural Text Mining with Large Language Models |
2023-10-05T07:57:09Z | http://arxiv.org/pdf/2310.03368v4 | http://arxiv.org/abs/2310.03368v4 | Qinyuan Cheng, Tianxiang Sun, Wenwei Zhang, Siyin Wang, Xiangyang Liu, Mozhi Zhang, Junliang He, Mianqiu Huang, Zhangyue Yin, Kai Chen, Xipeng Qiu | 2023-10-25T07:49:37Z | cs.CL | In this paper, we establish a benchmark named HalluQA (Chinese Hallucination
Question-Answering) to measure the hallucination phenomenon in Chinese large
language models. HalluQA contains 450 meticulously designed adversarial
questions, spanning multiple domains, and takes into account Chinese historical
culture, customs, and social phenomena. During the construction of HalluQA, we
consider two types of hallucinations: imitative falsehoods and factual errors,
and we construct adversarial samples based on GLM-130B and ChatGPT. For
evaluation, we design an automated evaluation method using GPT-4 to judge
whether a model output is hallucinated. We conduct extensive experiments on 24
large language models, including ERNIE-Bot, Baichuan2, ChatGLM, Qwen, SparkDesk
and etc. Out of the 24 models, 18 achieved non-hallucination rates lower than
50%. This indicates that HalluQA is highly challenging. We analyze the primary
types of hallucinations in different types of models and their causes.
Additionally, we discuss which types of hallucinations should be prioritized
for different types of models. | cs.CL | Evaluating Hallucinations in Chinese Large Language Models |
2023-10-05T06:31:01Z | http://arxiv.org/pdf/2310.04460v1 | http://arxiv.org/abs/2310.04460v1 | Jingyuan Sun, Xiaohan Zhang, Marie-Francine Moens | 2023-10-05T06:31:01Z | cs.CL, cs.AI | To understand the algorithm that supports the human brain's language
representation, previous research has attempted to predict neural responses to
linguistic stimuli using embeddings generated by artificial neural networks
(ANNs), a process known as neural encoding. However, most of these studies have
focused on probing neural representations of Germanic languages, such as
English, with unsupervised ANNs. In this paper, we propose to bridge the gap
between human brain and supervised ANN representations of the Chinese language.
Specifically, we investigate how task tuning influences a pretained Transformer
for neural encoding and which tasks lead to the best encoding performances. We
generate supervised representations on eight Natural Language Understanding
(NLU) tasks using prompt-tuning, a technique that is seldom explored in neural
encoding for language. We demonstrate that prompt-tuning yields representations
that better predict neural responses to Chinese stimuli than traditional
fine-tuning on four tasks. Furthermore, we discover that tasks that require a
fine-grained processing of concepts and entities lead to representations that
are most predictive of brain activation patterns. Additionally, we reveal that
the proportion of tuned parameters highly influences the neural encoding
performance of fine-tuned models. Overall, our experimental findings could help
us better understand the relationship between supervised artificial and brain
language representations. | cs.CL | Tuning In to Neural Encoding: Linking Human Brain and Artificial Supervised Representations of Language |
2023-10-05T05:55:06Z | http://arxiv.org/pdf/2310.03328v2 | http://arxiv.org/abs/2310.03328v2 | Zhen wan, Yating Zhang, Yexiang Wang, Fei Cheng, Sadao Kurohashi | 2023-10-12T22:14:52Z | cs.CL | While large language models (LLMs) like GPT-4 have recently demonstrated
astonishing zero-shot capabilities in general domain tasks, they often generate
content with hallucinations in specific domains such as Chinese law, hindering
their application in these areas. This is typically due to the absence of
training data that encompasses such a specific domain, preventing GPT-4 from
acquiring in-domain knowledge. A pressing challenge is that it's not plausible
to continue training LLMs of such scale on in-domain data.
This paper introduces a simple and effective domain adaptation framework for
GPT-4 by reformulating generation as an \textbf{adapt-retrieve-revise} process.
The initial step is to \textbf{adapt} an affordable 7B LLM to the target domain
by continuing learning on in-domain data. When solving a task, we leverage the
adapted LLM to generate a draft answer given a task query. Then, the draft
answer will be used to \textbf{retrieve} supporting evidence candidates from an
external in-domain knowledge base. Finally, the draft answer and retrieved
evidence are concatenated into a whole prompt to let GPT-4 assess the evidence
and \textbf{revise} the draft answer to generate the final answer.
Our proposal combines the advantages of the efficiency of adapting a smaller
7B model with the evidence-assessing capability of GPT-4 and effectively
prevents GPT-4 from generating hallucinatory content. In the zero-shot setting
of four Chinese legal tasks, our method improves accuracy by 33.3\% compared to
the direct generation by GPT-4. When compared to two stronger retrieval-based
baselines, our method outperforms them by 15.4\% and 23.9\%. Our code will be
released | cs.CL | Reformulating Domain Adaptation of Large Language Models as Adapt-Retrieve-Revise |
2023-10-05T04:47:49Z | http://arxiv.org/pdf/2310.03309v4 | http://arxiv.org/abs/2310.03309v4 | Junjie Liu, Shaotian Yan, Chen Shen, Liang Xie, Wenxiao Wang, Jieping Ye | 2024-06-13T06:26:46Z | cs.CL, cs.AI | Exploiting large language models (LLMs) to tackle reasoning has garnered
growing attention. It still remains highly challenging to achieve satisfactory
results in complex logical problems, characterized by plenty of premises within
the prompt and requiring multi-hop reasoning. In particular, the reasoning
capabilities of LLMs are brittle to disorder and distractibility. In this work,
we first examine the mechanism from the perspective of information flow and
reveal that LLMs exhibit failure patterns akin to human-like cognitive biases
when dealing with disordered and irrelevant content in reasoning tasks.
However, in contrast to LLMs, disordered and irrelevant content does not
significantly decrease human performance, as humans have a propensity to
distill the most relevant information and systematically organize their
thoughts, aiding them in responding to questions. Stem from that, we further
propose a novel reasoning approach named Concise and Organized Perception
(COP). COP carefully analyzes the given statements to identify the most
pertinent information while eliminating redundancy efficiently. It then prompts
the LLMs in a more organized form that adapts to the model's inference process.
By perceiving concise and organized context, the reasoning abilities of LLMs
can be better elicited. Extensive experimental results on several popular
logical benchmarks (ProofWriter, PrOntoQA, PrOntoQA-OOD, and FOLIO) and math
benchmark (DI-GSM) show that COP significantly outperforms previous
state-of-the-art methods. | cs.CL | Concise and Organized Perception Facilitates Reasoning in Large Language Models |
2023-10-05T04:15:48Z | http://arxiv.org/pdf/2310.03304v4 | http://arxiv.org/abs/2310.03304v4 | Danqing Wang, Kevin Yang, Hanlin Zhu, Xiaomeng Yang, Andrew Cohen, Lei Li, Yuandong Tian | 2024-06-19T22:05:12Z | cs.CL | With rapid progress made in language qualities such as fluency and
consistency via large language models (LLMs), there has been increasing
interest in assessing alignment with diverse human preferences. Traditional
metrics heavily rely on lexical similarity with human-written references and
have been observed to suffer from a poor correlation with human evaluation.
Furthermore, they ignore the diverse preferences of humans, a key aspect in
evaluating open-ended tasks like story generation. Inspired by these
challenges, we introduce an interpretable open-ended evaluation framework PerSE
to assess the alignment with a specific human preference. It is tuned to deduce
the specific preference from a given personal profile and evaluate the
alignment between the generation and the personal preference. PerSE also
explains its assessment by a detailed comment or several fine-grained scores.
This enhances its interpretability, making it more suitable to tailor a
personalized generation. Our 13B LLaMA-2-based PerSE shows a 15.8% increase in
Kendall correlation and a 13.7% rise in accuracy on zero-shot reviewers
compared to GPT-4. It also outperforms GPT-4 by 46.01% in the Kendall
correlation on new domains, indicating its transferability. | cs.CL | Learning Personalized Alignment for Evaluating Open-ended Text Generation |
2023-10-05T03:45:54Z | http://arxiv.org/pdf/2310.03293v1 | http://arxiv.org/abs/2310.03293v1 | Siwei Wu, Xiangqing Shen, Rui Xia | 2023-10-05T03:45:54Z | cs.CL | Large Language Models (LLMs), such as ChatGPT, have recently been applied to
various NLP tasks due to its open-domain generation capabilities. However,
there are two issues with applying LLMs to dialogue tasks. 1. During the
dialogue process, users may have implicit intentions that might be overlooked
by LLMs. Consequently, generated responses couldn't align with the user's
intentions. 2. It is unlikely for LLMs to encompass all fields comprehensively.
In certain specific domains, their knowledge may be incomplete, and LLMs cannot
update the latest knowledge in real-time. To tackle these issues, we propose a
framework~\emph{using LLM to \textbf{E}nhance dialogue response generation by
asking questions to \textbf{D}etect user's \textbf{I}mplicit
in\textbf{T}entions} (\textbf{EDIT}). Firstly, EDIT generates open questions
related to the dialogue context as the potential user's intention; Then, EDIT
answers those questions by interacting with LLMs and searching in
domain-specific knowledge bases respectively, and use LLMs to choose the proper
answers to questions as extra knowledge; Finally, EDIT enhances response
generation by explicitly integrating those extra knowledge. Besides, previous
question generation works only focus on asking questions with answers in
context. In order to ask open questions, we construct a Context-Open-Question
(COQ) dataset. On two task-oriented dialogue tasks (Wizard of Wikipedia and
Holl-E), EDIT outperformed other LLMs. | cs.CL | A New Dialogue Response Generation Agent for Large Language Models by Asking Questions to Detect User's Intentions |
2023-10-05T03:20:41Z | http://arxiv.org/pdf/2310.03283v1 | http://arxiv.org/abs/2310.03283v1 | Ke Shen, Mayank Kejriwal | 2023-10-05T03:20:41Z | cs.CL | Large Language Models (LLMs), such as ChatGPT, have achieved impressive
milestones in natural language processing (NLP). Despite their impressive
performance, the models are known to pose important risks. As these models are
deployed in real-world applications, a systematic understanding of different
risks posed by these models on tasks such as natural language inference (NLI),
is much needed. In this paper, we define and formalize two distinct types of
risk: decision risk and composite risk. We also propose a risk-centric
evaluation framework, and four novel metrics, for assessing LLMs on these risks
in both in-domain and out-of-domain settings. Finally, we propose a
risk-adjusted calibration method called DwD for helping LLMs minimize these
risks in an overall NLI architecture. Detailed experiments, using four NLI
benchmarks, three baselines and two LLMs, including ChatGPT, show both the
practical utility of the evaluation framework, and the efficacy of DwD in
reducing decision and composite risk. For instance, when using DwD, an
underlying LLM is able to address an extra 20.1% of low-risk inference tasks
(but which the LLM erroneously deems high-risk without risk adjustment) and
skip a further 19.8% of high-risk tasks, which would have been answered
incorrectly. | cs.CL | A Formalism and Approach for Improving Robustness of Large Language Models Using Risk-Adjusted Confidence Scores |
2023-10-05T02:51:51Z | http://arxiv.org/pdf/2310.03772v1 | http://arxiv.org/abs/2310.03772v1 | Neil Daniel | 2023-10-05T02:51:51Z | cs.CL, cs.LG | A common practice in the medical industry is the use of clinical notes, which
consist of detailed patient observations. However, electronic health record
systems frequently do not contain these observations in a structured format,
rendering patient information challenging to assess and evaluate automatically.
Using computational systems for the extraction of medical attributes offers
many applications, including longitudinal analysis of patients, risk
assessment, and hospital evaluation. Recent work has constructed successful
methods for phenotyping: extracting medical attributes from clinical notes.
BERT-based models can be used to transform clinical notes into a series of
representations, which are then condensed into a single document representation
based on their CLS embeddings and passed into an LSTM (Mulyar et al., 2020).
Though this pipeline yields a considerable performance improvement over
previous results, it requires extensive convergence time. This method also does
not allow for predicting attributes not yet identified in clinical notes.
Considering the wide variety of medical attributes that may be present in a
clinical note, we propose an alternative pipeline utilizing ScispaCy (Neumann
et al., 2019) for the extraction of common diseases. We then train various
supervised learning models to associate the presence of these conditions with
patient attributes. Finally, we replicate a ClinicalBERT (Alsentzer et al.,
2019) and LSTM-based approach for purposes of comparison. We find that
alternative methods moderately underperform the replicated LSTM approach. Yet,
considering a complex tradeoff between accuracy and runtime, in addition to the
fact that the alternative approach also allows for the detection of medical
conditions that are not already present in a clinical note, its usage may be
considered as a supplement to established methods. | cs.CL | Investigating Alternative Feature Extraction Pipelines For Clinical Note Phenotyping |
2023-10-05T02:45:39Z | http://arxiv.org/pdf/2310.03269v1 | http://arxiv.org/abs/2310.03269v1 | Zeyuan Wang, Qiang Zhang, Keyan Ding, Ming Qin, Xiang Zhuang, Xiaotong Li, Huajun Chen | 2023-10-05T02:45:39Z | q-bio.BM, cs.CL | Large Language Models (LLMs) have revolutionized the field of natural
language processing, but they fall short in comprehending biological sequences
such as proteins. To address this challenge, we propose InstructProtein, an
innovative LLM that possesses bidirectional generation capabilities in both
human and protein languages: (i) taking a protein sequence as input to predict
its textual function description and (ii) using natural language to prompt
protein sequence generation. To achieve this, we first pre-train an LLM on both
protein and natural language corpora, enabling it to comprehend individual
languages. Then supervised instruction tuning is employed to facilitate the
alignment of these two distinct languages. Herein, we introduce a knowledge
graph-based instruction generation framework to construct a high-quality
instruction dataset, addressing annotation imbalance and instruction deficits
in existing protein-text corpus. In particular, the instructions inherit the
structural relations between proteins and function annotations in knowledge
graphs, which empowers our model to engage in the causal modeling of protein
functions, akin to the chain-of-thought processes in natural languages.
Extensive experiments on bidirectional protein-text generation tasks show that
InstructProtein outperforms state-of-the-art LLMs by large margins. Moreover,
InstructProtein serves as a pioneering step towards text-based protein function
prediction and sequence design, effectively bridging the gap between protein
and human language understanding. | q-bio.BM | InstructProtein: Aligning Human and Protein Language via Knowledge Instruction |
2023-10-05T02:35:00Z | http://arxiv.org/pdf/2310.03262v3 | http://arxiv.org/abs/2310.03262v3 | Shengding Hu, Xin Liu, Xu Han, Xinrong Zhang, Chaoqun He, Weilin Zhao, Yankai Lin, Ning Ding, Zebin Ou, Guoyang Zeng, Zhiyuan Liu, Maosong Sun | 2024-04-17T13:43:59Z | cs.CL | The scientific scale-up of large language models (LLMs) necessitates a
comprehensive understanding of their scaling properties. However, the existing
literature on the scaling properties only yields an incomplete answer:
optimization loss decreases predictably as the model size increases, in line
with established scaling law; yet no scaling law for task has been established
and the task performances are far from predictable during scaling. Task
performances typically show minor gains on small models until they improve
dramatically once models exceed a size threshold, exemplifying the ``emergent
abilities''. In this study, we discover that small models, although they
exhibit minor performance, demonstrate critical and consistent task performance
improvements that are not captured by conventional evaluation strategies due to
insufficient measurement resolution. To measure such improvements, we introduce
PassUntil, an evaluation strategy with theoretically infinite resolution,
through massive sampling in the decoding phase. With PassUntil, we conduct a
quantitative investigation into the scaling law of task performance. The
investigation contains two parts. Firstly, a strict task scaling law that is
not conventionally known to exist, is identified, enhancing the predictability
of task performances. Remarkably, we are able to predict the performance of the
2.4B model on code generation with merely 0.05\% deviation before training
starts, which is the first systematic attempt to verify predictable scaling
proposed by GPT-4's report. Secondly, we are able to study emergent abilities
quantitatively. We identify a kind of accelerated emergence whose scaling curve
cannot be fitted by standard scaling law function and has a increasing speed.
We then examine two hypothesis and imply that the ``multiple circuits
hypothesis'' might be responsible for the accelerated emergence. | cs.CL | Predicting Emergent Abilities with Infinite Resolution Evaluation |
2023-10-05T01:42:16Z | http://arxiv.org/pdf/2310.03249v2 | http://arxiv.org/abs/2310.03249v2 | Mohamed Aghzal, Erion Plaku, Ziyu Yao | 2024-02-07T20:18:54Z | cs.CL | Large language models (LLMs) have achieved remarkable success across a wide
spectrum of tasks; however, they still face limitations in scenarios that
demand long-term planning and spatial reasoning. To facilitate this line of
research, in this work, we propose a new benchmark, termed $\textbf{P}$ath
$\textbf{P}$lanning from $\textbf{N}$atural $\textbf{L}$anguage
($\textbf{PPNL}$). Our benchmark evaluates LLMs' spatial-temporal reasoning by
formulating ''path planning'' tasks that require an LLM to navigate to target
locations while avoiding obstacles and adhering to constraints. Leveraging this
benchmark, we systematically investigate LLMs including GPT-4 via different
few-shot prompting methodologies as well as BART and T5 of various sizes via
fine-tuning. Our experimental results show the promise of few-shot GPT-4 in
spatial reasoning, when it is prompted to reason and act interleavedly,
although it still fails to perform long-term temporal reasoning. In contrast,
while fine-tuned LLMs achieved impressive results on in-distribution reasoning
tasks, they struggled to generalize to larger environments or environments with
more obstacles. | cs.CL | Can Large Language Models be Good Path Planners? A Benchmark and Investigation on Spatial-temporal Reasoning |
2023-10-05T00:55:46Z | http://arxiv.org/pdf/2310.03232v1 | http://arxiv.org/abs/2310.03232v1 | Xinyang Ren, Hannah A Burkhardt, Patricia A Areán, Thomas D Hull, Trevor Cohen | 2023-10-05T00:55:46Z | cs.CL, cs.AI | Prior work has shown that analyzing the use of first-person singular pronouns
can provide insight into individuals' mental status, especially depression
symptom severity. These findings were generated by counting frequencies of
first-person singular pronouns in text data. However, counting doesn't capture
how these pronouns are used. Recent advances in neural language modeling have
leveraged methods generating contextual embeddings. In this study, we sought to
utilize the embeddings of first-person pronouns obtained from contextualized
language representation models to capture ways these pronouns are used, to
analyze mental status. De-identified text messages sent during online
psychotherapy with weekly assessment of depression severity were used for
evaluation. Results indicate the advantage of contextualized first-person
pronoun embeddings over standard classification token embeddings and
frequency-based pronoun analysis results in predicting depression symptom
severity. This suggests contextual representations of first-person pronouns can
enhance the predictive utility of language used by people with depression
symptoms. | cs.CL | Deep Representations of First-person Pronouns for Prediction of Depression Symptom Severity |
2023-10-05T00:04:12Z | http://arxiv.org/pdf/2310.03214v2 | http://arxiv.org/abs/2310.03214v2 | Tu Vu, Mohit Iyyer, Xuezhi Wang, Noah Constant, Jerry Wei, Jason Wei, Chris Tar, Yun-Hsuan Sung, Denny Zhou, Quoc Le, Thang Luong | 2023-11-22T07:28:19Z | cs.CL | Most large language models (LLMs) are trained once and never updated; thus,
they lack the ability to dynamically adapt to our ever-changing world. In this
work, we perform a detailed study of the factuality of LLM-generated text in
the context of answering questions that test current world knowledge.
Specifically, we introduce FreshQA, a novel dynamic QA benchmark encompassing a
diverse range of question and answer types, including questions that require
fast-changing world knowledge as well as questions with false premises that
need to be debunked. We benchmark a diverse array of both closed and
open-source LLMs under a two-mode evaluation procedure that allows us to
measure both correctness and hallucination. Through human evaluations involving
more than 50K judgments, we shed light on limitations of these models and
demonstrate significant room for improvement: for instance, all models
(regardless of model size) struggle on questions that involve fast-changing
knowledge and false premises. Motivated by these results, we present
FreshPrompt, a simple few-shot prompting method that substantially boosts the
performance of an LLM on FreshQA by incorporating relevant and up-to-date
information retrieved from a search engine into the prompt. Our experiments
show that FreshPrompt outperforms both competing search engine-augmented
prompting methods such as Self-Ask (Press et al., 2022) as well as commercial
systems such as Perplexity.AI. Further analysis of FreshPrompt reveals that
both the number of retrieved evidences and their order play a key role in
influencing the correctness of LLM-generated answers. Additionally, instructing
the LLM to generate concise and direct answers helps reduce hallucination
compared to encouraging more verbose answers. To facilitate future work, we
release FreshQA at github.com/freshllms/freshqa and commit to updating it at
regular intervals. | cs.CL | FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation |
2023-10-04T23:33:36Z | http://arxiv.org/pdf/2310.03211v2 | http://arxiv.org/abs/2310.03211v2 | Utsav Garg, Erhan Bas | 2023-11-28T03:50:54Z | cs.CL, cs.AI, cs.CV | Instruction-tuned large language models (LLMs) have demonstrated promising
zero-shot generalization capabilities across various downstream tasks. Recent
research has introduced multimodal capabilities to LLMs by integrating
independently pretrained vision encoders through model grafting. These
multimodal variants undergo instruction tuning, similar to LLMs, enabling
effective zero-shot generalization for multimodal tasks. This study conducts a
comparative analysis of different multimodal instruction tuning approaches and
evaluates their performance across a range of tasks, including complex
reasoning, conversation, image captioning, multiple-choice questions (MCQs),
and binary classification. Through rigorous benchmarking and ablation
experiments, we reveal key insights for guiding architectural choices when
incorporating multimodal capabilities into LLMs. However, current approaches
have limitations; they do not sufficiently address the need for a diverse
multimodal instruction dataset, which is crucial for enhancing task
generalization. Additionally, they overlook issues related to truthfulness and
factuality when generating responses. These findings illuminate current
methodological constraints in adapting language models for image comprehension
and provide valuable guidance for researchers and practitioners seeking to
harness multimodal versions of LLMs. | cs.CL | On the Performance of Multimodal Language Models |
2023-10-04T23:32:33Z | http://arxiv.org/pdf/2310.03210v1 | http://arxiv.org/abs/2310.03210v1 | Erfan Al-Hossami, Razvan Bunescu, Justin Smith, Ryan Teehan | 2023-10-04T23:32:33Z | cs.CL, cs.CY | When employing the Socratic method of teaching, instructors guide students
toward solving a problem on their own rather than providing the solution
directly. While this strategy can substantially improve learning outcomes, it
is usually time-consuming and cognitively demanding. Automated Socratic
conversational agents can augment human instruction and provide the necessary
scale, however their development is hampered by the lack of suitable data for
training and evaluation. In this paper, we introduce a manually created dataset
of multi-turn Socratic advice that is aimed at helping a novice programmer fix
buggy solutions to simple computational problems. The dataset is then used for
benchmarking the Socratic debugging abilities of a number of language models,
ranging from fine-tuning the instruction-based text-to-text transformer Flan-T5
to zero-shot and chain of thought prompting of the much larger GPT-4. The code
and datasets are made freely available for research at the link below.
https://github.com/taisazero/socratic-debugging-benchmark | cs.CL | Can Language Models Employ the Socratic Method? Experiments with Code Debugging |
2023-10-04T22:34:56Z | http://arxiv.org/pdf/2310.03193v1 | http://arxiv.org/abs/2310.03193v1 | Hancheng Cao, Jesse Dodge, Kyle Lo, Daniel A. McFarland, Lucy Lu Wang | 2023-10-04T22:34:56Z | cs.DL, cs.CL, cs.CY, physics.hist-ph, physics.soc-ph | In recent years, funding agencies and journals increasingly advocate for open
science practices (e.g. data and method sharing) to improve the transparency,
access, and reproducibility of science. However, quantifying these practices at
scale has proven difficult. In this work, we leverage a large-scale dataset of
1.1M papers from arXiv that are representative of the fields of physics, math,
and computer science to analyze the adoption of data and method link-sharing
practices over time and their impact on article reception. To identify links to
data and methods, we train a neural text classification model to automatically
classify URL types based on contextual mentions in papers. We find evidence
that the practice of link-sharing to methods and data is spreading as more
papers include such URLs over time. Reproducibility efforts may also be
spreading because the same links are being increasingly reused across papers
(especially in computer science); and these links are increasingly concentrated
within fewer web domains (e.g. Github) over time. Lastly, articles that share
data and method links receive increased recognition in terms of citation count,
with a stronger effect when the shared links are active (rather than defunct).
Together, these findings demonstrate the increased spread and perceived value
of data and method sharing practices in open science. | cs.DL | The Rise of Open Science: Tracking the Evolution and Perceived Value of Data and Methods Link-Sharing Practices |
2023-10-04T22:09:28Z | http://arxiv.org/pdf/2310.03184v2 | http://arxiv.org/abs/2310.03184v2 | Zachary Levonian, Chenglu Li, Wangda Zhu, Anoushka Gade, Owen Henkel, Millie-Ellen Postle, Wanli Xing | 2023-11-11T01:26:22Z | cs.CL, cs.HC | For middle-school math students, interactive question-answering (QA) with
tutors is an effective way to learn. The flexibility and emergent capabilities
of generative large language models (LLMs) has led to a surge of interest in
automating portions of the tutoring process - including interactive QA to
support conceptual discussion of mathematical concepts. However, LLM responses
to math questions can be incorrect or mismatched to the educational context -
such as being misaligned with a school's curriculum. One potential solution is
retrieval-augmented generation (RAG), which involves incorporating a vetted
external knowledge source in the LLM prompt to increase response quality. In
this paper, we designed prompts that retrieve and use content from a
high-quality open-source math textbook to generate responses to real student
questions. We evaluate the efficacy of this RAG system for middle-school
algebra and geometry QA by administering a multi-condition survey, finding that
humans prefer responses generated using RAG, but not when responses are too
grounded in the textbook content. We argue that while RAG is able to improve
response quality, designers of math QA systems must consider trade-offs between
generating responses preferred by students and responses closely matched to
specific educational resources. | cs.CL | Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preference |
2023-10-04T21:57:09Z | http://arxiv.org/pdf/2310.03182v1 | http://arxiv.org/abs/2310.03182v1 | An Yan, Yu Wang, Yiwu Zhong, Zexue He, Petros Karypis, Zihan Wang, Chengyu Dong, Amilcare Gentili, Chun-Nan Hsu, Jingbo Shang, Julian McAuley | 2023-10-04T21:57:09Z | cs.CV, cs.CL, cs.LG | Medical image classification is a critical problem for healthcare, with the
potential to alleviate the workload of doctors and facilitate diagnoses of
patients. However, two challenges arise when deploying deep learning models to
real-world healthcare applications. First, neural models tend to learn spurious
correlations instead of desired features, which could fall short when
generalizing to new domains (e.g., patients with different ages). Second, these
black-box models lack interpretability. When making diagnostic predictions, it
is important to understand why a model makes a decision for trustworthy and
safety considerations. In this paper, to address these two limitations, we
propose a new paradigm to build robust and interpretable medical image
classifiers with natural language concepts. Specifically, we first query
clinical concepts from GPT-4, then transform latent image features into
explicit concepts with a vision-language model. We systematically evaluate our
method on eight medical image classification datasets to verify its
effectiveness. On challenging datasets with strong confounding factors, our
method can mitigate spurious correlations thus substantially outperform
standard visual encoders and other baselines. Finally, we show how
classification with a small number of concepts brings a level of
interpretability for understanding model decisions through case studies in real
medical data. | cs.CV | Robust and Interpretable Medical Image Classifiers via Concept Bottleneck Models |
2023-10-04T21:40:36Z | http://arxiv.org/pdf/2310.03173v2 | http://arxiv.org/abs/2310.03173v2 | Zishun Yu, Yunzhe Tao, Liyu Chen, Tao Sun, Hongxia Yang | 2024-03-18T21:44:47Z | cs.CL | Program synthesis aims to create accurate, executable programs from problem
specifications, specifically from natural language descriptions in our context.
Recent studies have leveraged the power of reinforcement learning (RL) in
conjunction with large language models (LLMs), significantly enhancing code
generation capabilities. The application of RL focuses on directly optimizing
for functional correctness, offering an advantage over conventional supervised
methods. Despite policy-based RL methods dominating the literature on RL for
program synthesis, the nature of program synthesis tasks hints at a natural
alignment with value-based methods. This stems from the rich collection of
off-policy programs, including those developed by human programmers and also
historical samples, coupled with the straightforward verification of generated
programs through automated unit testing, meaning rewards are easy to obtain.
Diverging from the dominant use of policy-based algorithms, our work explores
the feasibility of value-based approaches, leading to the development of our
$\mathcal{B}$-Coder (pronounced Bellman coder). Yet, training value-based
methods presents challenges due to the enormous search space inherent to
program synthesis. To this end, we introduce an initialization protocol for RL
agents utilizing pre-trained LMs and a conservative Bellman operator to reduce
training complexities. Moreover, we demonstrate how to leverage the learned
value functions as a dual strategy to post-process generated programs. Our
empirical evaluations demonstrated $\mathcal{B}$-Coder's capability in
achieving state-of-the-art performance when compared to policy-based methods.
Remarkably, this achievement is reached with minimal reward engineering effort,
highlighting the effectiveness of value-based RL, independent of reward
designs. | cs.CL | $\mathcal{B}$-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis |
2023-10-04T19:39:26Z | http://arxiv.org/pdf/2310.03128v5 | http://arxiv.org/abs/2310.03128v5 | Yue Huang, Jiawen Shi, Yuan Li, Chenrui Fan, Siyuan Wu, Qihui Zhang, Yixin Liu, Pan Zhou, Yao Wan, Neil Zhenqiang Gong, Lichao Sun | 2024-02-23T13:19:52Z | cs.SE, cs.CL | Large language models (LLMs) have garnered significant attention due to their
impressive natural language processing (NLP) capabilities. Recently, many
studies have focused on the tool utilization ability of LLMs. They primarily
investigated how LLMs effectively collaborate with given specific tools.
However, in scenarios where LLMs serve as intelligent agents, as seen in
applications like AutoGPT and MetaGPT, LLMs are expected to engage in intricate
decision-making processes that involve deciding whether to employ a tool and
selecting the most suitable tool(s) from a collection of available tools to
fulfill user requests. Therefore, in this paper, we introduce MetaTool, a
benchmark designed to evaluate whether LLMs have tool usage awareness and can
correctly choose tools. Specifically, we create a dataset called ToolE within
the benchmark. This dataset contains various types of user queries in the form
of prompts that trigger LLMs to use tools, including both single-tool and
multi-tool scenarios. Subsequently, we set the tasks for both tool usage
awareness and tool selection. We define four subtasks from different
perspectives in tool selection, including tool selection with similar choices,
tool selection in specific scenarios, tool selection with possible reliability
issues, and multi-tool selection. We conduct experiments involving eight
popular LLMs and find that the majority of them still struggle to effectively
select tools, highlighting the existing gaps between LLMs and genuine
intelligent agents. However, through the error analysis, we found there is
still significant room for improvement. Finally, we conclude with insights for
tool developers -- we strongly recommend that tool developers choose an
appropriate rewrite model for generating new descriptions based on the
downstream LLM the tool will apply to. Our code is in
https://github.com/HowieHwong/MetaTool. | cs.SE | MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use |
2023-10-04T18:21:17Z | http://arxiv.org/pdf/2310.03094v3 | http://arxiv.org/abs/2310.03094v3 | Murong Yue, Jie Zhao, Min Zhang, Liang Du, Ziyu Yao | 2024-02-08T22:02:22Z | cs.CL, cs.AI, cs.LG | Large language models (LLMs) such as GPT-4 have exhibited remarkable
performance in a variety of tasks, but this strong performance often comes with
the high expense of using paid API services. In this paper, we are motivated to
study building an LLM cascade to save the cost of using LLMs, particularly for
performing reasoning (e.g., mathematical, causal) tasks. Our cascade pipeline
follows the intuition that simpler questions can be addressed by a weaker but
more affordable LLM, whereas only the challenging questions necessitate the
stronger and more expensive LLM. To realize this decision-making, we consider
the "answer consistency" of the weaker LLM as a signal of the question
difficulty and propose several methods for the answer sampling and consistency
checking, including one leveraging a mixture of two thought representations
(i.e., Chain-of-Thought and Program-of-Thought). Through experiments on six
reasoning benchmark datasets, with GPT-3.5-turbo and GPT-4 being the weaker and
stronger LLMs, respectively, we demonstrate that our proposed LLM cascades can
achieve performance comparable to using solely the stronger LLM but require
only 40% of its cost. | cs.CL | Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning |
2023-10-04T18:02:01Z | http://arxiv.org/pdf/2310.03084v1 | http://arxiv.org/abs/2310.03084v1 | Deniz Bayazit, Negar Foroutan, Zeming Chen, Gail Weiss, Antoine Bosselut | 2023-10-04T18:02:01Z | cs.CL, cs.AI, cs.LG | Pretrained language models (LMs) encode implicit representations of knowledge
in their parameters. However, localizing these representations and
disentangling them from each other remains an open problem. In this work, we
investigate whether pretrained language models contain various
knowledge-critical subnetworks: particular sparse computational subgraphs
responsible for encoding specific knowledge the model has memorized. We propose
a multi-objective differentiable weight masking scheme to discover these
subnetworks and show that we can use them to precisely remove specific
knowledge from models while minimizing adverse effects on the behavior of the
original language model. We demonstrate our method on multiple GPT2 variants,
uncovering highly sparse subnetworks (98%+) that are solely responsible for
specific collections of relational knowledge. When these subnetworks are
removed, the remaining network maintains most of its initial capacity (modeling
language and other memorized relational knowledge) but struggles to express the
removed knowledge, and suffers performance drops on examples needing this
removed knowledge on downstream tasks after finetuning. | cs.CL | Discovering Knowledge-Critical Subnetworks in Pretrained Language Models |
2023-10-04T17:59:49Z | http://arxiv.org/pdf/2310.03026v2 | http://arxiv.org/abs/2310.03026v2 | Hao Sha, Yao Mu, Yuxuan Jiang, Li Chen, Chenfeng Xu, Ping Luo, Shengbo Eben Li, Masayoshi Tomizuka, Wei Zhan, Mingyu Ding | 2023-10-13T16:13:43Z | cs.RO, cs.AI, cs.CL, cs.CV, cs.LG | Existing learning-based autonomous driving (AD) systems face challenges in
comprehending high-level information, generalizing to rare events, and
providing interpretability. To address these problems, this work employs Large
Language Models (LLMs) as a decision-making component for complex AD scenarios
that require human commonsense understanding. We devise cognitive pathways to
enable comprehensive reasoning with LLMs, and develop algorithms for
translating LLM decisions into actionable driving commands. Through this
approach, LLM decisions are seamlessly integrated with low-level controllers by
guided parameter matrix adaptation. Extensive experiments demonstrate that our
proposed method not only consistently surpasses baseline approaches in
single-vehicle tasks, but also helps handle complex driving behaviors even
multi-vehicle coordination, thanks to the commonsense reasoning capabilities of
LLMs. This paper presents an initial step toward leveraging LLMs as effective
decision-makers for intricate AD scenarios in terms of safety, efficiency,
generalizability, and interoperability. We aspire for it to serve as
inspiration for future research in this field. Project page:
https://sites.google.com/view/llm-mpc | cs.RO | LanguageMPC: Large Language Models as Decision Makers for Autonomous Driving |
2023-10-04T17:59:41Z | http://arxiv.org/pdf/2310.03025v2 | http://arxiv.org/abs/2310.03025v2 | Peng Xu, Wei Ping, Xianchao Wu, Lawrence McAfee, Chen Zhu, Zihan Liu, Sandeep Subramanian, Evelina Bakhturina, Mohammad Shoeybi, Bryan Catanzaro | 2024-01-23T07:49:13Z | cs.CL, cs.AI, cs.IR, cs.LG | Extending the context window of large language models (LLMs) is getting
popular recently, while the solution of augmenting LLMs with retrieval has
existed for years. The natural questions are: i) Retrieval-augmentation versus
long context window, which one is better for downstream tasks? ii) Can both
methods be combined to get the best of both worlds? In this work, we answer
these questions by studying both solutions using two state-of-the-art
pretrained LLMs, i.e., a proprietary 43B GPT and Llama2-70B. Perhaps
surprisingly, we find that LLM with 4K context window using simple
retrieval-augmentation at generation can achieve comparable performance to
finetuned LLM with 16K context window via positional interpolation on long
context tasks, while taking much less computation. More importantly, we
demonstrate that retrieval can significantly improve the performance of LLMs
regardless of their extended context window sizes. Our best model,
retrieval-augmented Llama2-70B with 32K context window, outperforms
GPT-3.5-turbo-16k and Davinci003 in terms of average score on nine long context
tasks including question answering, query-based summarization, and in-context
few-shot learning tasks. It also outperforms its non-retrieval Llama2-70B-32k
baseline by a margin, while being much faster at generation. Our study provides
general insights on the choice of retrieval-augmentation versus long context
extension of LLM for practitioners. | cs.CL | Retrieval meets Long Context Large Language Models |
2023-10-04T17:58:11Z | http://arxiv.org/pdf/2310.03018v3 | http://arxiv.org/abs/2310.03018v3 | Kuan-Po Huang, Chih-Kai Yang, Yu-Kuan Fu, Ewan Dunbar, Hung-yi Lee | 2024-03-18T07:57:58Z | eess.AS, cs.CL, cs.SD | We introduce a new zero resource code-switched speech benchmark designed to
directly assess the code-switching capabilities of self-supervised speech
encoders. We showcase a baseline system of language modeling on discrete units
to demonstrate how the code-switching abilities of speech encoders can be
assessed in a zero-resource manner. Our experiments encompass a variety of
well-known speech encoders, including Wav2vec 2.0, HuBERT, XLSR, etc. We
examine the impact of pre-training languages and model size on benchmark
performance. Notably, though our results demonstrate that speech encoders with
multilingual pre-training, exemplified by XLSR, outperform monolingual variants
(Wav2vec 2.0, HuBERT) in code-switching scenarios, there is still substantial
room for improvement in their code-switching linguistic abilities. | eess.AS | Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages |
2023-10-04T17:58:05Z | http://arxiv.org/pdf/2310.03017v1 | http://arxiv.org/abs/2310.03017v1 | Yuxuan Sun, Kai Zhang, Yu Su | 2023-10-04T17:58:05Z | cs.CL | Multimodal information extraction (MIE) aims to extract structured
information from unstructured multimedia content. Due to the diversity of tasks
and settings, most current MIE models are task-specific and data-intensive,
which limits their generalization to real-world scenarios with diverse task
requirements and limited labeled data. To address these issues, we propose a
novel multimodal question answering (MQA) framework to unify three MIE tasks by
reformulating them into a unified span extraction and multi-choice QA pipeline.
Extensive experiments on six datasets show that: 1) Our MQA framework
consistently and significantly improves the performances of various
off-the-shelf large multimodal models (LMM) on MIE tasks, compared to vanilla
prompting. 2) In the zero-shot setting, MQA outperforms previous
state-of-the-art baselines by a large margin. In addition, the effectiveness of
our framework can successfully transfer to the few-shot setting, enhancing LMMs
on a scale of 10B parameters to be competitive or outperform much larger
language models such as ChatGPT and GPT-4. Our MQA framework can serve as a
general principle of utilizing LMMs to better solve MIE and potentially other
downstream multimodal tasks. | cs.CL | Multimodal Question Answering for Unified Information Extraction |
2023-10-04T17:57:33Z | http://arxiv.org/pdf/2310.03016v1 | http://arxiv.org/abs/2310.03016v1 | Satwik Bhattamishra, Arkil Patel, Phil Blunsom, Varun Kanade | 2023-10-04T17:57:33Z | cs.LG, cs.CL | In order to understand the in-context learning phenomenon, recent works have
adopted a stylized experimental framework and demonstrated that Transformers
can learn gradient-based learning algorithms for various classes of real-valued
functions. However, the limitations of Transformers in implementing learning
algorithms, and their ability to learn other forms of algorithms are not well
understood. Additionally, the degree to which these capabilities are confined
to attention-based models is unclear. Furthermore, it remains to be seen
whether the insights derived from these stylized settings can be extrapolated
to pretrained Large Language Models (LLMs). In this work, we take a step
towards answering these questions by demonstrating the following: (a) On a
test-bed with a variety of Boolean function classes, we find that Transformers
can nearly match the optimal learning algorithm for 'simpler' tasks, while
their performance deteriorates on more 'complex' tasks. Additionally, we find
that certain attention-free models perform (almost) identically to Transformers
on a range of tasks. (b) When provided a teaching sequence, i.e. a set of
examples that uniquely identifies a function in a class, we show that
Transformers learn more sample-efficiently. Interestingly, our results show
that Transformers can learn to implement two distinct algorithms to solve a
single task, and can adaptively select the more sample-efficient algorithm
depending on the sequence of in-context examples. (c) Lastly, we show that
extant LLMs, e.g. LLaMA-2, GPT-4, can compete with nearest-neighbor baselines
on prediction tasks that are guaranteed to not be in their training set. | cs.LG | Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions |
2023-10-04T17:41:59Z | http://arxiv.org/pdf/2310.03003v1 | http://arxiv.org/abs/2310.03003v1 | Siddharth Samsi, Dan Zhao, Joseph McDonald, Baolin Li, Adam Michaleas, Michael Jones, William Bergeron, Jeremy Kepner, Devesh Tiwari, Vijay Gadepally | 2023-10-04T17:41:59Z | cs.CL, cs.DC | Large language models (LLMs) have exploded in popularity due to their new
generative capabilities that go far beyond prior state-of-the-art. These
technologies are increasingly being leveraged in various domains such as law,
finance, and medicine. However, these models carry significant computational
challenges, especially the compute and energy costs required for inference.
Inference energy costs already receive less attention than the energy costs of
training LLMs -- despite how often these large models are called on to conduct
inference in reality (e.g., ChatGPT). As these state-of-the-art LLMs see
increasing usage and deployment in various domains, a better understanding of
their resource utilization is crucial for cost-savings, scaling performance,
efficient hardware usage, and optimal inference strategies.
In this paper, we describe experiments conducted to study the computational
and energy utilization of inference with LLMs. We benchmark and conduct a
preliminary analysis of the inference performance and inference energy costs of
different sizes of LLaMA -- a recent state-of-the-art LLM -- developed by Meta
AI on two generations of popular GPUs (NVIDIA V100 \& A100) and two datasets
(Alpaca and GSM8K) to reflect the diverse set of tasks/benchmarks for LLMs in
research and practice. We present the results of multi-node, multi-GPU
inference using model sharding across up to 32 GPUs. To our knowledge, our work
is the one of the first to study LLM inference performance from the perspective
of computational and energy resources at this scale. | cs.CL | From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference |
2023-10-04T17:34:00Z | http://arxiv.org/pdf/2310.02998v2 | http://arxiv.org/abs/2310.02998v2 | Yi-Lin Sung, Jaehong Yoon, Mohit Bansal | 2024-01-26T18:45:29Z | cs.CV, cs.AI, cs.CL, cs.LG | Large Vision-Language Models (LVLMs) can understand the world comprehensively
by integrating rich information from different modalities, achieving remarkable
advancements on various multimodal downstream tasks. However, deploying LVLMs
is often problematic due to their massive computational/energy costs and carbon
consumption. Such issues make it infeasible to adopt conventional iterative
global pruning, which is costly due to computing the Hessian matrix of the
entire large model for sparsification. Alternatively, several studies have
recently proposed layer-wise pruning approaches to avoid the expensive
computation of global pruning and efficiently compress model weights according
to their importance within a layer. However, they often suffer from suboptimal
model compression due to their lack of a global perspective. To address this
limitation in recent efficient pruning methods for large models, we propose
Efficient Coarse-to-Fine LayerWise Pruning (ECoFLaP), a two-stage
coarse-to-fine weight pruning approach for LVLMs. We first determine the
sparsity ratios of different layers or blocks by leveraging the global
importance score, which is efficiently computed based on the zeroth-order
approximation of the global model gradients. Then, the model performs local
layer-wise unstructured weight pruning based on globally-informed sparsity
ratios. We validate our proposed method across various multimodal and unimodal
models and datasets, demonstrating significant performance improvements over
prevalent pruning techniques in the high-sparsity regime. | cs.CV | ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models |
2023-10-04T17:28:44Z | http://arxiv.org/pdf/2310.02992v3 | http://arxiv.org/abs/2310.02992v3 | Xichen Pan, Li Dong, Shaohan Huang, Zhiliang Peng, Wenhu Chen, Furu Wei | 2024-04-26T01:24:57Z | cs.CV, cs.CL | Recent advancements in subject-driven image generation have made significant
strides. However, current methods still fall short in diverse application
scenarios, as they require test-time tuning and cannot accept interleaved
multi-image and text input. These limitations keep them far from the ultimate
goal of "image as a foreign language in image generation." This paper presents
Kosmos-G, a model that leverages the advanced multimodal perception
capabilities of Multimodal Large Language Models (MLLMs) to tackle the
aforementioned challenge. Our approach aligns the output space of MLLM with
CLIP using the textual modality as an anchor and performs compositional
instruction tuning on curated data. Kosmos-G demonstrates an impressive
capability of zero-shot subject-driven generation with interleaved multi-image
and text input. Notably, the score distillation instruction tuning requires no
modifications to the image decoder. This allows for a seamless substitution of
CLIP and effortless integration with a myriad of U-Net techniques ranging from
fine-grained controls to personalized image decoder variants. We posit Kosmos-G
as an initial attempt towards the goal of "image as a foreign language in image
generation." The code can be found at https://aka.ms/Kosmos-G | cs.CV | Kosmos-G: Generating Images in Context with Multimodal Large Language Models |
2023-10-04T17:26:16Z | http://arxiv.org/pdf/2310.02989v1 | http://arxiv.org/abs/2310.02989v1 | Siavash Golkar, Mariel Pettee, Michael Eickenberg, Alberto Bietti, Miles Cranmer, Geraud Krawezik, Francois Lanusse, Michael McCabe, Ruben Ohana, Liam Parker, Bruno Régaldo-Saint Blancard, Tiberiu Tesileanu, Kyunghyun Cho, Shirley Ho | 2023-10-04T17:26:16Z | stat.ML, cs.AI, cs.CL, cs.LG | Large Language Models have not yet been broadly adapted for the analysis of
scientific datasets due in part to the unique difficulties of tokenizing
numbers. We propose xVal, a numerical encoding scheme that represents any real
number using just a single token. xVal represents a given real number by
scaling a dedicated embedding vector by the number value. Combined with a
modified number-inference approach, this strategy renders the model end-to-end
continuous when considered as a map from the numbers of the input string to
those of the output string. This leads to an inductive bias that is generally
more suitable for applications in scientific domains. We empirically evaluate
our proposal on a number of synthetic and real-world datasets. Compared with
existing number encoding schemes, we find that xVal is more token-efficient and
demonstrates improved generalization. | stat.ML | xVal: A Continuous Number Encoding for Large Language Models |
2023-10-04T17:20:34Z | http://arxiv.org/pdf/2310.02984v2 | http://arxiv.org/abs/2310.02984v2 | Vivien Cabannes, Elvis Dohmatob, Alberto Bietti | 2024-02-20T19:17:52Z | stat.ML, cs.AI, cs.CL, cs.LG, cs.NE, I.2.6; G.1.6 | Learning arguably involves the discovery and memorization of abstract rules.
The aim of this paper is to study associative memory mechanisms. Our model is
based on high-dimensional matrices consisting of outer products of embeddings,
which relates to the inner layers of transformer language models. We derive
precise scaling laws with respect to sample size and parameter size, and
discuss the statistical efficiency of different estimators, including
optimization-based algorithms. We provide extensive numerical experiments to
validate and interpret theoretical results, including fine-grained
visualizations of the stored memory associations. | stat.ML | Scaling Laws for Associative Memories |
2023-10-04T17:17:06Z | http://arxiv.org/pdf/2310.02980v4 | http://arxiv.org/abs/2310.02980v4 | Ido Amos, Jonathan Berant, Ankit Gupta | 2024-04-28T13:52:55Z | cs.LG, cs.CL | Modeling long-range dependencies across sequences is a longstanding goal in
machine learning and has led to architectures, such as state space models, that
dramatically outperform Transformers on long sequences. However, these
impressive empirical gains have been by and large demonstrated on benchmarks
(e.g. Long Range Arena), where models are randomly initialized and trained to
predict a target label from an input sequence. In this work, we show that
random initialization leads to gross overestimation of the differences between
architectures and that pretraining with standard denoising objectives, using
$\textit{only the downstream task data}$, leads to dramatic gains across
multiple architectures and to very small gaps between Transformers and state
space models (SSMs). In stark contrast to prior works, we find vanilla
Transformers to match the performance of S4 on Long Range Arena when properly
pretrained, and we improve the best reported results of SSMs on the PathX-256
task by 20 absolute points. Subsequently, we analyze the utility of
previously-proposed structured parameterizations for SSMs and show they become
mostly redundant in the presence of data-driven initialization obtained through
pretraining. Our work shows that, when evaluating different architectures on
supervised tasks, incorporation of data-driven priors via pretraining is
essential for reliable performance estimation, and can be done efficiently. | cs.LG | Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors |
2023-10-04T17:12:18Z | http://arxiv.org/pdf/2310.02977v2 | http://arxiv.org/abs/2310.02977v2 | Yuze He, Yushi Bai, Matthieu Lin, Wang Zhao, Yubin Hu, Jenny Sheng, Ran Yi, Juanzi Li, Yong-Jin Liu | 2024-04-17T09:09:17Z | cs.CV, cs.CL, cs.LG | Recent methods in text-to-3D leverage powerful pretrained diffusion models to
optimize NeRF. Notably, these methods are able to produce high-quality 3D
scenes without training on 3D data. Due to the open-ended nature of the task,
most studies evaluate their results with subjective case studies and user
experiments, thereby presenting a challenge in quantitatively addressing the
question: How has current progress in Text-to-3D gone so far? In this paper, we
introduce T$^3$Bench, the first comprehensive text-to-3D benchmark containing
diverse text prompts of three increasing complexity levels that are specially
designed for 3D generation. To assess both the subjective quality and the text
alignment, we propose two automatic metrics based on multi-view images produced
by the 3D contents. The quality metric combines multi-view text-image scores
and regional convolution to detect quality and view inconsistency. The
alignment metric uses multi-view captioning and GPT-4 evaluation to measure
text-3D consistency. Both metrics closely correlate with different dimensions
of human judgments, providing a paradigm for efficiently evaluating text-to-3D
models. The benchmarking results, shown in Fig. 1, reveal performance
differences among an extensive 10 prevalent text-to-3D methods. Our analysis
further highlights the common struggles for current methods on generating
surroundings and multi-object scenes, as well as the bottleneck of leveraging
2D guidance for 3D generation. Our project page is available at:
https://t3bench.com. | cs.CV | T$^3$Bench: Benchmarking Current Progress in Text-to-3D Generation |
2023-10-04T17:10:23Z | http://arxiv.org/pdf/2310.02973v2 | http://arxiv.org/abs/2310.02973v2 | Siddhant Arora, Hayato Futami, Jee-weon Jung, Yifan Peng, Roshan Sharma, Yosuke Kashiwagi, Emiru Tsunoo, Karen Livescu, Shinji Watanabe | 2024-04-03T14:12:36Z | cs.CL, cs.SD, eess.AS | Recent studies leverage large language models with multi-tasking
capabilities, using natural language prompts to guide the model's behavior and
surpassing performance of task-specific models. Motivated by this, we ask: can
we build a single model that jointly performs various spoken language
understanding (SLU) tasks? We start by adapting a pre-trained automatic speech
recognition model to additional tasks using single-token task specifiers. We
enhance this approach through instruction tuning, i.e., finetuning by
describing the task using natural language instructions followed by the list of
label options. Our approach can generalize to new task descriptions for the
seen tasks during inference, thereby enhancing its user-friendliness. We
demonstrate the efficacy of our single multi-task learning model "UniverSLU"
for 12 speech classification and sequence generation task types spanning 17
datasets and 9 languages. On most tasks, UniverSLU achieves competitive
performance and often even surpasses task-specific models. Additionally, we
assess the zero-shot capabilities, finding that the model generalizes to new
datasets and languages for seen task types. | cs.CL | UniverSLU: Universal Spoken Language Understanding for Diverse Tasks with Natural Language Instructions |
2023-10-04T17:07:32Z | http://arxiv.org/pdf/2310.02971v3 | http://arxiv.org/abs/2310.02971v3 | Kai-Wei Chang, Ming-Hsin Chen, Yun-Ping Lin, Jing Neng Hsu, Paul Kuo-Ming Huang, Chien-yu Huang, Shang-Wen Li, Hung-yi Lee | 2023-11-14T21:15:01Z | eess.AS, cs.CL, eess.SP | Prompting and adapter tuning have emerged as efficient alternatives to
fine-tuning (FT) methods. However, existing studies on speech prompting focused
on classification tasks and failed on more complex sequence generation tasks.
Besides, adapter tuning is primarily applied with a focus on encoder-only
self-supervised models. Our experiments show that prompting on Wav2Seq, a
self-supervised encoder-decoder model, surpasses previous works in sequence
generation tasks. It achieves a remarkable 53% relative improvement in word
error rate for ASR and a 27% in F1 score for slot filling. Additionally,
prompting competes with the FT method in the low-resource scenario. Moreover,
we show the transferability of prompting and adapter tuning on Wav2Seq in
cross-lingual ASR. When limited trainable parameters are involved, prompting
and adapter tuning consistently outperform conventional FT across 7 languages.
Notably, in the low-resource scenario, prompting consistently outperforms
adapter tuning. | eess.AS | Prompting and Adapter Tuning for Self-supervised Encoder-Decoder Speech Model |
2023-10-04T16:44:37Z | http://arxiv.org/pdf/2310.02954v5 | http://arxiv.org/abs/2310.02954v5 | Jing Xiong, Zixuan Li, Chuanyang Zheng, Zhijiang Guo, Yichun Yin, Enze Xie, Zhicheng Yang, Qingxing Cao, Haiming Wang, Xiongwei Han, Jing Tang, Chengming Li, Xiaodan Liang | 2024-03-02T14:38:24Z | cs.CL | Recent advances in natural language processing, primarily propelled by Large
Language Models (LLMs), have showcased their remarkable capabilities grounded
in in-context learning. A promising avenue for guiding LLMs in intricate
reasoning tasks involves the utilization of intermediate reasoning steps within
the Chain-of-Thought (CoT) paradigm. Nevertheless, the central challenge lies
in the effective selection of exemplars for facilitating in-context learning.
In this study, we introduce a framework that leverages Dual Queries and
Low-rank approximation Re-ranking (DQ-LoRe) to automatically select exemplars
for in-context learning. Dual Queries first query LLM to obtain LLM-generated
knowledge such as CoT, then query the retriever to obtain the final exemplars
via both question and the knowledge. Moreover, for the second query, LoRe
employs dimensionality reduction techniques to refine exemplar selection,
ensuring close alignment with the input question's knowledge. Through extensive
experiments, we demonstrate that DQ-LoRe significantly outperforms prior
state-of-the-art methods in the automatic selection of exemplars for GPT-4,
enhancing performance from 92.5% to 94.2%. Our comprehensive analysis further
reveals that DQ-LoRe consistently outperforms retrieval-based approaches in
terms of both performance and adaptability, especially in scenarios
characterized by distribution shifts. DQ-LoRe pushes the boundary of in-context
learning and opens up new avenues for addressing complex reasoning challenges.
Our code is released at
https://github.com/AI4fun/DQ-LoRe}{https://github.com/AI4fun/DQ-LoRe. | cs.CL | DQ-LoRe: Dual Queries with Low Rank Approximation Re-ranking for In-Context Learning |
2023-10-04T16:44:23Z | http://arxiv.org/pdf/2310.02953v3 | http://arxiv.org/abs/2310.02953v3 | Chang Gao, Wenxuan Zhang, Guizhen Chen, Wai Lam | 2024-05-24T13:44:12Z | cs.CL | Instruction tuning has become an essential process for optimizing the
performance of large language models (LLMs). However, current text-to-text
instruction tuning methods, referred to as TextTuning, exhibit significant
limitations in terms of generalization, robustness, and controllability,
primarily due to the absence of explicit task structures. In this paper, we
introduce JsonTuning, a novel structure-to-structure approach for instruction
tuning. By utilizing the versatile and structured format of JSON to represent
tasks, JsonTuning enhances generalization by enabling the model to comprehend
essential task elements and their interrelations, improves robustness by
reducing ambiguity, and increases controllability by providing explicit control
over the output. We conduct a comprehensive comparative analysis between
JsonTuning and TextTuning using various language models and evaluation
benchmarks. Our experimental results demonstrate that JsonTuning consistently
outperforms TextTuning across a range of applications, showing marked
improvements in performance, robustness, and controllability. By addressing the
inherent limitations of TextTuning, JsonTuning reveals significant potential
for developing more effective and reliable LLMs capable of managing diverse
scenarios. | cs.CL | JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning |
2023-10-04T16:39:31Z | http://arxiv.org/pdf/2310.02949v1 | http://arxiv.org/abs/2310.02949v1 | Xianjun Yang, Xiao Wang, Qi Zhang, Linda Petzold, William Yang Wang, Xun Zhao, Dahua Lin | 2023-10-04T16:39:31Z | cs.CL, cs.AI, cs.CR, cs.LG | Warning: This paper contains examples of harmful language, and reader
discretion is recommended. The increasing open release of powerful large
language models (LLMs) has facilitated the development of downstream
applications by reducing the essential cost of data annotation and computation.
To ensure AI safety, extensive safety-alignment measures have been conducted to
armor these models against malicious use (primarily hard prompt attack).
However, beneath the seemingly resilient facade of the armor, there might lurk
a shadow. By simply tuning on 100 malicious examples with 1 GPU hour, these
safely aligned LLMs can be easily subverted to generate harmful content.
Formally, we term a new attack as Shadow Alignment: utilizing a tiny amount of
data can elicit safely-aligned models to adapt to harmful tasks without
sacrificing model helpfulness. Remarkably, the subverted models retain their
capability to respond appropriately to regular inquiries. Experiments across 8
models released by 5 different organizations (LLaMa-2, Falcon, InternLM,
BaiChuan2, Vicuna) demonstrate the effectiveness of shadow alignment attack.
Besides, the single-turn English-only attack successfully transfers to
multi-turn dialogue and other languages. This study serves as a clarion call
for a collective effort to overhaul and fortify the safety of open-source LLMs
against malicious attackers. | cs.CL | Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models |
2023-10-04T16:37:05Z | http://arxiv.org/pdf/2310.12321v1 | http://arxiv.org/abs/2310.12321v1 | Katikapalli Subramanyam Kalyan | 2023-10-04T16:37:05Z | cs.CL | Large language models (LLMs) are a special class of pretrained language
models obtained by scaling model size, pretraining corpus and computation.
LLMs, because of their large size and pretraining on large volumes of text
data, exhibit special abilities which allow them to achieve remarkable
performances without any task-specific training in many of the natural language
processing tasks. The era of LLMs started with OpenAI GPT-3 model, and the
popularity of LLMs is increasing exponentially after the introduction of models
like ChatGPT and GPT4. We refer to GPT-3 and its successor OpenAI models,
including ChatGPT and GPT4, as GPT-3 family large language models (GLLMs). With
the ever-rising popularity of GLLMs, especially in the research community,
there is a strong need for a comprehensive survey which summarizes the recent
research progress in multiple dimensions and can guide the research community
with insightful future research directions. We start the survey paper with
foundation concepts like transformers, transfer learning, self-supervised
learning, pretrained language models and large language models. We then present
a brief overview of GLLMs and discuss the performances of GLLMs in various
downstream tasks, specific domains and multiple languages. We also discuss the
data labelling and data augmentation abilities of GLLMs, the robustness of
GLLMs, the effectiveness of GLLMs as evaluators, and finally, conclude with
multiple insightful future research directions. To summarize, this
comprehensive survey paper will serve as a good resource for both academic and
industry people to stay updated with the latest research related to GPT-3
family large language models. | cs.CL | A Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4 |
2023-10-04T16:23:37Z | http://arxiv.org/pdf/2310.02943v1 | http://arxiv.org/abs/2310.02943v1 | Aleksandr Meister, Matvei Novikov, Nikolay Karpov, Evelina Bakhturina, Vitaly Lavrukhin, Boris Ginsburg | 2023-10-04T16:23:37Z | cs.CL | Traditional automatic speech recognition (ASR) models output lower-cased
words without punctuation marks, which reduces readability and necessitates a
subsequent text processing model to convert ASR transcripts into a proper
format. Simultaneously, the development of end-to-end ASR models capable of
predicting punctuation and capitalization presents several challenges,
primarily due to limited data availability and shortcomings in the existing
evaluation methods, such as inadequate assessment of punctuation prediction. In
this paper, we introduce a LibriSpeech-PC benchmark designed to assess the
punctuation and capitalization prediction capabilities of end-to-end ASR
models. The benchmark includes a LibriSpeech-PC dataset with restored
punctuation and capitalization, a novel evaluation metric called Punctuation
Error Rate (PER) that focuses on punctuation marks, and initial baseline
models. All code, data, and models are publicly available. | cs.CL | LibriSpeech-PC: Benchmark for Evaluation of Punctuation and Capitalization Capabilities of end-to-end ASR Models |
2023-10-04T16:09:48Z | http://arxiv.org/pdf/2310.02932v2 | http://arxiv.org/abs/2310.02932v2 | Jannis Bulian, Mike S. Schäfer, Afra Amini, Heidi Lam, Massimiliano Ciaramita, Ben Gaiarin, Michelle Chen Hübscher, Christian Buck, Niels G. Mede, Markus Leippold, Nadine Strauß | 2024-05-28T15:36:49Z | cs.CL, cs.AI, cs.CY, cs.LG | As Large Language Models (LLMs) rise in popularity, it is necessary to assess
their capability in critically relevant domains. We present a comprehensive
evaluation framework, grounded in science communication research, to assess LLM
responses to questions about climate change. Our framework emphasizes both
presentational and epistemological adequacy, offering a fine-grained analysis
of LLM generations spanning 8 dimensions and 30 issues. Our evaluation task is
a real-world example of a growing number of challenging problems where AI can
complement and lift human performance. We introduce a novel protocol for
scalable oversight that relies on AI Assistance and raters with relevant
education. We evaluate several recent LLMs on a set of diverse climate
questions. Our results point to a significant gap between surface and
epistemological qualities of LLMs in the realm of climate communication. | cs.CL | Assessing Large Language Models on Climate Information |
2023-10-04T15:10:06Z | http://arxiv.org/pdf/2310.02876v1 | http://arxiv.org/abs/2310.02876v1 | Aman Khullar, Daniel Nkemelu, Cuong V. Nguyen, Michael L. Best | 2023-10-04T15:10:06Z | cs.CL, cs.CY | A growing body of work has focused on text classification methods for
detecting the increasing amount of hate speech posted online. This progress has
been limited to only a select number of highly-resourced languages causing
detection systems to either under-perform or not exist in limited data
contexts. This is majorly caused by a lack of training data which is expensive
to collect and curate in these settings. In this work, we propose a data
augmentation approach that addresses the problem of lack of data for online
hate speech detection in limited data contexts using synthetic data generation
techniques. Given a handful of hate speech examples in a high-resource language
such as English, we present three methods to synthesize new examples of hate
speech data in a target language that retains the hate sentiment in the
original examples but transfers the hate targets. We apply our approach to
generate training data for hate speech classification tasks in Hindi and
Vietnamese. Our findings show that a model trained on synthetic data performs
comparably to, and in some cases outperforms, a model trained only on the
samples available in the target domain. This method can be adopted to bootstrap
hate speech detection models from scratch in limited data contexts. As the
growth of social media within these contexts continues to outstrip response
efforts, this work furthers our capacities for detection, understanding, and
response to hate speech. | cs.CL | Hate Speech Detection in Limited Data Contexts using Synthetic Data Generation |
2023-10-04T14:11:12Z | http://arxiv.org/pdf/2310.02842v2 | http://arxiv.org/abs/2310.02842v2 | Chen Dun, Mirian Hipolito Garcia, Guoqing Zheng, Ahmed Hassan Awadallah, Anastasios Kyrillidis, Robert Sim | 2023-10-05T21:13:11Z | cs.CL, cs.AI | Large Language Models (LLMs) have the ability to solve a variety of tasks,
such as text summarization and mathematical questions, just out of the box, but
they are often trained with a single task in mind. Due to high computational
costs, the current trend is to use prompt instruction tuning to better adjust
monolithic, pretrained LLMs for new -- but often individual -- downstream
tasks. Thus, how one would expand prompt tuning to handle -- concomitantly --
heterogeneous tasks and data distributions is a widely open question. To
address this gap, we suggest the use of \emph{Mixture of Prompts}, or MoPs,
associated with smart gating functionality: the latter -- whose design is one
of the contributions of this paper -- can identify relevant skills embedded in
different groups of prompts and dynamically assign combined experts (i.e.,
collection of prompts), based on the target task. Additionally, MoPs are
empirically agnostic to any model compression technique applied -- for
efficiency reasons -- as well as instruction data source and task composition.
In practice, MoPs can simultaneously mitigate prompt training "interference" in
multi-task, multi-source scenarios (e.g., task and data heterogeneity across
sources), as well as possible implications from model approximations. As a
highlight, MoPs manage to decrease final perplexity from $\sim20\%$ up to
$\sim70\%$, as compared to baselines, in the federated scenario, and from $\sim
3\%$ up to $\sim30\%$ in the centralized scenario. | cs.CL | Sweeping Heterogeneity with Smart MoPs: Mixture of Prompts for LLM Task Adaptation |
2023-10-04T13:59:45Z | http://arxiv.org/pdf/2310.02832v2 | http://arxiv.org/abs/2310.02832v2 | Fran Jelenić, Josip Jukić, Martin Tutek, Mate Puljiz, Jan Šnajder | 2024-03-11T18:18:41Z | cs.LG, cs.CL | Effective out-of-distribution (OOD) detection is crucial for reliable machine
learning models, yet most current methods are limited in practical use due to
requirements like access to training data or intervention in training. We
present a novel method for detecting OOD data in Transformers based on
transformation smoothness between intermediate layers of a network (BLOOD),
which is applicable to pre-trained models without access to training data.
BLOOD utilizes the tendency of between-layer representation transformations of
in-distribution (ID) data to be smoother than the corresponding transformations
of OOD data, a property that we also demonstrate empirically. We evaluate BLOOD
on several text classification tasks with Transformer networks and demonstrate
that it outperforms methods with comparable resource requirements. Our analysis
also suggests that when learning simpler tasks, OOD data transformations
maintain their original sharpness, whereas sharpness increases with more
complex tasks. | cs.LG | Out-of-Distribution Detection by Leveraging Between-Layer Transformation Smoothness |
2023-10-04T13:54:46Z | http://arxiv.org/pdf/2310.04456v1 | http://arxiv.org/abs/2310.04456v1 | Shihao Zou, Xianying Huang, Xudong Shen | 2023-10-04T13:54:46Z | cs.CL, cs.SD, eess.AS | Emotion Recognition in Conversation (ERC) plays an important role in driving
the development of human-machine interaction. Emotions can exist in multiple
modalities, and multimodal ERC mainly faces two problems: (1) the noise problem
in the cross-modal information fusion process, and (2) the prediction problem
of less sample emotion labels that are semantically similar but different
categories. To address these issues and fully utilize the features of each
modality, we adopted the following strategies: first, deep emotion cues
extraction was performed on modalities with strong representation ability, and
feature filters were designed as multimodal prompt information for modalities
with weak representation ability. Then, we designed a Multimodal Prompt
Transformer (MPT) to perform cross-modal information fusion. MPT embeds
multimodal fusion information into each attention layer of the Transformer,
allowing prompt information to participate in encoding textual features and
being fused with multi-level textual information to obtain better multimodal
fusion features. Finally, we used the Hybrid Contrastive Learning (HCL)
strategy to optimize the model's ability to handle labels with few samples.
This strategy uses unsupervised contrastive learning to improve the
representation ability of multimodal fusion and supervised contrastive learning
to mine the information of labels with few samples. Experimental results show
that our proposed model outperforms state-of-the-art models in ERC on two
benchmark datasets. | cs.CL | Multimodal Prompt Transformer with Hybrid Contrastive Learning for Emotion Recognition in Conversation |
2023-10-04T13:29:47Z | http://arxiv.org/pdf/2310.02804v1 | http://arxiv.org/abs/2310.02804v1 | Peifang Wang, Olga Golovneva, Armen Aghajanyan, Xiang Ren, Muhao Chen, Asli Celikyilmaz, Maryam Fazel-Zarandi | 2023-10-04T13:29:47Z | cs.CL, cs.CV, cs.LG | Visual language reasoning requires a system to extract text or numbers from
information-dense images like charts or plots and perform logical or arithmetic
reasoning to arrive at an answer. To tackle this task, existing work relies on
either (1) an end-to-end vision-language model trained on a large amount of
data, or (2) a two-stage pipeline where a captioning model converts the image
into text that is further read by another large language model to deduce the
answer. However, the former approach forces the model to answer a complex
question with one single step, and the latter approach is prone to inaccurate
or distracting information in the converted text that can confuse the language
model. In this work, we propose a dual-system for multi-step multimodal
reasoning, which consists of a "System-1" step for visual information
extraction and a "System-2" step for deliberate reasoning. Given an input,
System-2 breaks down the question into atomic sub-steps, each guiding System-1
to extract the information required for reasoning from the image. Experiments
on chart and plot datasets show that our method with a pre-trained System-2
module performs competitively compared to prior work on in- and
out-of-distribution data. By fine-tuning the System-2 module (LLaMA-2 70B) on
only a small amount of data on multi-step reasoning, the accuracy of our method
is further improved and surpasses the best fully-supervised end-to-end approach
by 5.7% and a pipeline approach with FlanPaLM (540B) by 7.5% on a challenging
dataset with human-authored questions. | cs.CL | DOMINO: A Dual-System for Multi-step Visual Language Reasoning |
2023-10-04T13:09:39Z | http://arxiv.org/pdf/2310.02790v1 | http://arxiv.org/abs/2310.02790v1 | Mubashir Munaf, Hammad Afzal, Naima Iltaf, Khawir Mahmood | 2023-10-04T13:09:39Z | cs.CL | With the advent of Deep Learning based Artificial Neural Networks models,
Natural Language Processing (NLP) has witnessed significant improvements in
textual data processing in terms of its efficiency and accuracy. However, the
research is mostly restricted to high-resource languages such as English and
low-resource languages still suffer from a lack of available resources in terms
of training datasets as well as models with even baseline evaluation results.
Considering the limited availability of resources for low-resource languages,
we propose a methodology for adapting self-attentive transformer-based
architecture models (mBERT, mT5) for low-resource summarization, supplemented
by the construction of a new baseline dataset (76.5k article, summary pairs) in
a low-resource language Urdu. Choosing news (a publicly available source) as
the application domain has the potential to make the proposed methodology
useful for reproducing in other languages with limited resources. Our adapted
summarization model \textit{urT5} with up to 44.78\% reduction in size as
compared to \textit{mT5} can capture contextual information of low resource
language effectively with evaluation score (up to 46.35 ROUGE-1, 77 BERTScore)
at par with state-of-the-art models in high resource language English
\textit{(PEGASUS: 47.21, BART: 45.14 on XSUM Dataset)}. The proposed method
provided a baseline approach towards extractive as well as abstractive
summarization with competitive evaluation results in a limited resource setup. | cs.CL | Low Resource Summarization using Pre-trained Language Models |
2023-10-04T12:50:26Z | http://arxiv.org/pdf/2310.02778v2 | http://arxiv.org/abs/2310.02778v2 | Rui Yang, Edison Marrese-Taylor, Yuhe Ke, Lechao Cheng, Qingyu Chen, Irene Li | 2023-10-13T12:10:01Z | cs.CL, cs.AI | Large language models (LLMs) have demonstrated powerful text generation
capabilities, bringing unprecedented innovation to the healthcare field. While
LLMs hold immense promise for applications in healthcare, applying them to real
clinical scenarios presents significant challenges, as these models may
generate content that deviates from established medical facts and even exhibit
potential biases. In our research, we develop an augmented LLM framework based
on the Unified Medical Language System (UMLS), aiming to better serve the
healthcare community. We employ LLaMa2-13b-chat and ChatGPT-3.5 as our
benchmark models, and conduct automatic evaluations using the ROUGE Score and
BERTScore on 104 questions from the LiveQA test set. Additionally, we establish
criteria for physician-evaluation based on four dimensions: Factuality,
Completeness, Readability and Relevancy. ChatGPT-3.5 is used for physician
evaluation with 20 questions on the LiveQA test set. Multiple resident
physicians conducted blind reviews to evaluate the generated content, and the
results indicate that this framework effectively enhances the factuality,
completeness, and relevance of generated content. Our research demonstrates the
effectiveness of using UMLS-augmented LLMs and highlights the potential
application value of LLMs in in medical question-answering. | cs.CL | Integrating UMLS Knowledge into Large Language Models for Medical Question Answering |
2023-10-04T12:48:33Z | http://arxiv.org/pdf/2310.02777v1 | http://arxiv.org/abs/2310.02777v1 | Chenwei Wu, Li Erran Li, Stefano Ermon, Patrick Haffner, Rong Ge, Zaiwei Zhang | 2023-10-04T12:48:33Z | cs.CL | Compositionality is a common property in many modalities including natural
languages and images, but the compositional generalization of multi-modal
models is not well-understood. In this paper, we identify two sources of
visual-linguistic compositionality: linguistic priors and the interplay between
images and texts. We show that current attempts to improve compositional
generalization rely on linguistic priors rather than on information in the
image. We also propose a new metric for compositionality without such
linguistic priors. | cs.CL | The Role of Linguistic Priors in Measuring Compositional Generalization of Vision-Language Models |
2023-10-04T12:14:43Z | http://arxiv.org/pdf/2310.02759v1 | http://arxiv.org/abs/2310.02759v1 | Bagiya Lakshmi S, Sanjjushri Varshini R, Rohith Mahadevan, Raja CSP Raman | 2023-10-04T12:14:43Z | cs.LG, cs.CL | Automated Essay Score (AES) is proven to be one of the cutting-edge
technologies. Scoring techniques are used for various purposes. Reliable scores
are calculated based on influential variables. Such variables can be computed
by different methods based on the domain. The research is concentrated on the
user's understanding of a given topic. The analysis is based on a scoring index
by using Large Language Models. The user can then compare and contrast the
understanding of a topic that they recently learned. The results are then
contributed towards learning analytics and progression is made for enhancing
the learning ability. In this research, the focus is on summarizing a PDF
document and gauging a user's understanding of its content. The process
involves utilizing a Langchain tool to summarize the PDF and extract the
essential information. By employing this technique, the research aims to
determine how well the user comprehends the summarized content. | cs.LG | Comparative Study and Framework for Automated Summariser Evaluation: LangChain and Hybrid Algorithms |
2023-10-04T11:49:37Z | http://arxiv.org/pdf/2310.02754v2 | http://arxiv.org/abs/2310.02754v2 | Paul Tardy, Charlotte Roze, Paul Poupet | 2023-10-05T14:28:20Z | cs.CL | Being able to read and understand written text is critical in a digital era.
However, studies shows that a large fraction of the population experiences
comprehension issues. In this context, further initiatives in accessibility are
required to improve the audience text comprehension. However, writers are
hardly assisted nor encouraged to produce easy-to-understand content. Moreover,
Automatic Text Simplification (ATS) model development suffers from the lack of
metric to accurately estimate comprehension difficulty We present
\textsc{LC-Score}, a simple approach for training text comprehension metric for
any French text without reference \ie predicting how easy to understand a given
text is on a $[0, 100]$ scale. Our objective with this scale is to
quantitatively capture the extend to which a text suits to the \textit{Langage
Clair} (LC, \textit{Clear Language}) guidelines, a French initiative closely
related to English Plain Language. We explore two approaches: (i) using
linguistically motivated indicators used to train statistical models, and (ii)
neural learning directly from text leveraging pre-trained language models. We
introduce a simple proxy task for comprehension difficulty training as a
classification task. To evaluate our models, we run two distinct human
annotation experiments, and find that both approaches (indicator based and
neural) outperforms commonly used readability and comprehension metrics such as
FKGL. | cs.CL | LC-Score: Reference-less estimation of Text Comprehension Difficulty |
2023-10-04T11:20:28Z | http://arxiv.org/pdf/2310.04455v1 | http://arxiv.org/abs/2310.04455v1 | Zihao Zhao, Zhenpeng Shi, Yang Liu, Wenbo Ding | 2023-10-04T11:20:28Z | cs.LG, cs.AI, cs.CL | Federated Learning (FL) is often impeded by communication overhead issues.
Prompt tuning, as a potential solution, has been introduced to only adjust a
few trainable parameters rather than the whole model. However, current
single-modality prompt tuning approaches fail to comprehensively portray local
clients' data. To overcome this limitation, we present Twin Prompt Federated
learning (TPFL), a pioneering solution that integrates both visual and textual
modalities, ensuring a more holistic representation of local clients' data
characteristics. Furthermore, in order to tackle the data heterogeneity issues,
we introduce the Augmented TPFL (ATPFL) employing the contrastive learning to
TPFL, which not only enhances the global knowledge acquisition of client models
but also fosters the development of robust, compact models. The effectiveness
of TPFL and ATPFL is substantiated by our extensive evaluations, consistently
showing superior performance compared to all baselines. | cs.LG | Inclusive Data Representation in Federated Learning: A Novel Approach Integrating Textual and Visual Prompt |
2023-10-04T09:28:59Z | http://arxiv.org/pdf/2310.04454v1 | http://arxiv.org/abs/2310.04454v1 | Eren Unlu | 2023-10-04T09:28:59Z | cs.CL, cs.AI | Position encoding is the primary mechanism which induces notion of sequential
order for input tokens in transformer architectures. Even though this
formulation in the original transformer paper has yielded plausible performance
for general purpose language understanding and generation, several new
frameworks such as Rotary Position Embedding (RoPE) are proposed for further
enhancement. In this paper, we introduce the notion of "geotokens" which are
input elements for transformer architectures, each representing an information
related to a geological location. Unlike the natural language the sequential
position is not important for the model but the geographical coordinates are.
In order to induce the concept of relative position for such a setting and
maintain the proportion between the physical distance and distance on embedding
space, we formulate a position encoding mechanism based on RoPE architecture
which is adjusted for spherical coordinates. | cs.CL | Spherical Position Encoding for Transformers |
2023-10-04T08:30:22Z | http://arxiv.org/pdf/2310.04453v1 | http://arxiv.org/abs/2310.04453v1 | Nicholas Perikli, Srimoy Bhattacharya, Blessing Ogbuokiri, Zahra Movahedi Nia, Benjamin Lieberman, Nidhi Tripathi, Salah-Eddine Dahbi, Finn Stevenson, Nicola Bragazzi, Jude Kong, Bruce Mellado | 2023-10-04T08:30:22Z | cs.CL, cs.LG, cs.SI | Very large numbers of M-pox cases have, since the start of May 2022, been
reported in non-endemic countries leading many to fear that the M-pox Outbreak
would rapidly transition into another pandemic, while the COVID-19 pandemic
ravages on. Given the similarities of M-pox with COVID-19, we chose to test the
performance of COVID-19 models trained on South African twitter data on a
hand-labelled M-pox dataset before and after fine-tuning. More than 20k
M-pox-related tweets from South Africa were hand-labelled as being either
positive, negative or neutral. After fine-tuning these COVID-19 models on the
M-pox dataset, the F1-scores increased by more than 8% falling just short of
70%, but still outperforming state-of-the-art models and well-known
classification algorithms. An LDA-based topic modelling procedure was used to
compare the miss-classified M-pox tweets of the original COVID-19 RoBERTa model
with its fine-tuned version, and from this analysis, we were able to draw
conclusions on how to build more sophisticated models. | cs.CL | COVID-19 South African Vaccine Hesitancy Models Show Boost in Performance Upon Fine-Tuning on M-pox Tweets |
2023-10-04T08:25:37Z | http://arxiv.org/pdf/2310.02655v1 | http://arxiv.org/abs/2310.02655v1 | Filippo Perrina, Francesco Marchiori, Mauro Conti, Nino Vincenzo Verde | 2023-10-04T08:25:37Z | cs.CR, cs.CL | Cyber Threat Intelligence (CTI) reporting is pivotal in contemporary risk
management strategies. As the volume of CTI reports continues to surge, the
demand for automated tools to streamline report generation becomes increasingly
apparent. While Natural Language Processing techniques have shown potential in
handling text data, they often struggle to address the complexity of diverse
data sources and their intricate interrelationships. Moreover, established
paradigms like STIX have emerged as de facto standards within the CTI
community, emphasizing the formal categorization of entities and relations to
facilitate consistent data sharing. In this paper, we introduce AGIR (Automatic
Generation of Intelligence Reports), a transformative Natural Language
Generation tool specifically designed to address the pressing challenges in the
realm of CTI reporting. AGIR's primary objective is to empower security
analysts by automating the labor-intensive task of generating comprehensive
intelligence reports from formal representations of entity graphs. AGIR
utilizes a two-stage pipeline by combining the advantages of template-based
approaches and the capabilities of Large Language Models such as ChatGPT. We
evaluate AGIR's report generation capabilities both quantitatively and
qualitatively. The generated reports accurately convey information expressed
through formal language, achieving a high recall value (0.99) without
introducing hallucination. Furthermore, we compare the fluency and utility of
the reports with state-of-the-art approaches, showing how AGIR achieves higher
scores in terms of Syntactic Log-Odds Ratio (SLOR) and through questionnaires.
By using our tool, we estimate that the report writing time is reduced by more
than 40%, therefore streamlining the CTI production of any organization and
contributing to the automation of several CTI tasks. | cs.CR | AGIR: Automating Cyber Threat Intelligence Reporting with Natural Language Generation |
2023-10-04T06:47:58Z | http://arxiv.org/pdf/2310.03051v1 | http://arxiv.org/abs/2310.03051v1 | Pei Zhou, Aman Madaan, Srividya Pranavi Potharaju, Aditya Gupta, Kevin R. McKee, Ari Holtzman, Jay Pujara, Xiang Ren, Swaroop Mishra, Aida Nematzadeh, Shyam Upadhyay, Manaal Faruqui | 2023-10-04T06:47:58Z | cs.CL, cs.AI | "Thinking is for Doing." Humans can infer other people's mental states from
observations--an ability called Theory-of-Mind (ToM)--and subsequently act
pragmatically on those inferences. Existing question answering benchmarks such
as ToMi ask models questions to make inferences about beliefs of characters in
a story, but do not test whether models can then use these inferences to guide
their actions. We propose a new evaluation paradigm for large language models
(LLMs): Thinking for Doing (T4D), which requires models to connect inferences
about others' mental states to actions in social scenarios. Experiments on T4D
demonstrate that LLMs such as GPT-4 and PaLM 2 seemingly excel at tracking
characters' beliefs in stories, but they struggle to translate this capability
into strategic action. Our analysis reveals the core challenge for LLMs lies in
identifying the implicit inferences about mental states without being
explicitly asked about as in ToMi, that lead to choosing the correct action in
T4D. To bridge this gap, we introduce a zero-shot prompting framework, Foresee
and Reflect (FaR), which provides a reasoning structure that encourages LLMs to
anticipate future challenges and reason about potential actions. FaR boosts
GPT-4's performance from 50% to 71% on T4D, outperforming other prompting
methods such as Chain-of-Thought and Self-Ask. Moreover, FaR generalizes to
diverse out-of-distribution story structures and scenarios that also require
ToM inferences to choose an action, consistently outperforming other methods
including few-shot in-context learning. | cs.CL | How FaR Are Large Language Models From Agents with Theory-of-Mind? |
2023-10-04T05:45:23Z | http://arxiv.org/pdf/2310.02594v1 | http://arxiv.org/abs/2310.02594v1 | Tianjun Mao, Chenghong Zhang | 2023-10-04T05:45:23Z | cs.CL | Spoken language understanding (SLU) typically includes two subtasks: intent
detection and slot filling. Currently, it has achieved great success in
high-resource languages, but it still remains challenging in low-resource
languages due to the scarcity of labeled training data. Hence, there is a
growing interest in zero-shot cross-lingual SLU. Despite of the success of
existing zero-shot cross-lingual SLU models, most of them neglect to achieve
the mutual guidance between intent and slots. To address this issue, we propose
an Intra-Inter Knowledge Distillation framework for zero-shot cross-lingual
Spoken Language Understanding (I$^2$KD-SLU) to model the mutual guidance.
Specifically, we not only apply intra-knowledge distillation between intent
predictions or slot predictions of the same utterance in different languages,
but also apply inter-knowledge distillation between intent predictions and slot
predictions of the same utterance. Our experimental results demonstrate that
our proposed framework significantly improves the performance compared with the
strong baselines and achieves the new state-of-the-art performance on the
MultiATIS++ dataset, obtaining a significant improvement over the previous best
model in overall accuracy. | cs.CL | I$^2$KD-SLU: An Intra-Inter Knowledge Distillation Framework for Zero-Shot Cross-Lingual Spoken Language Understanding |
2023-10-04T03:59:57Z | http://arxiv.org/pdf/2310.02567v2 | http://arxiv.org/abs/2310.02567v2 | Oscar Mañas, Benno Krojer, Aishwarya Agrawal | 2024-01-10T17:00:05Z | cs.CV, cs.AI, cs.CL, cs.LG | 8 years after the visual question answering (VQA) task was proposed, accuracy
remains the primary metric for automatic evaluation. VQA Accuracy has been
effective so far in the IID evaluation setting. However, our community is
undergoing a shift towards open-ended generative models and OOD evaluation. In
this new paradigm, the existing VQA Accuracy metric is overly stringent and
underestimates the performance of VQA systems. Thus, there is a need to develop
more robust automatic VQA metrics that serve as a proxy for human judgment. In
this work, we propose to leverage the in-context learning capabilities of
instruction-tuned large language models (LLMs) to build a better VQA metric. We
formulate VQA evaluation as an answer-rating task where the LLM is instructed
to score the accuracy of a candidate answer given a set of reference answers.
We demonstrate the proposed metric better correlates with human judgment
compared to existing metrics across several VQA models and benchmarks. We hope
wide adoption of our metric will contribute to better estimating the research
progress on the VQA task. We plan to release the evaluation code and collected
human judgments. | cs.CV | Improving Automatic VQA Evaluation Using Large Language Models |
2023-10-04T03:30:24Z | http://arxiv.org/pdf/2310.02556v2 | http://arxiv.org/abs/2310.02556v2 | Soroush Abbasi Koohpayegani, KL Navaneet, Parsa Nooralinejad, Soheil Kolouri, Hamed Pirsiavash | 2024-04-30T01:58:16Z | cs.CL, cs.CV | Fine-tuning Large Language Models (LLMs) and storing them for each downstream
task or domain is impractical because of the massive model size (e.g., 350GB in
GPT-3). Current literature, such as LoRA, showcases the potential of low-rank
modifications to the original weights of an LLM, enabling efficient adaptation
and storage for task-specific models. These methods can reduce the number of
parameters needed to fine-tune an LLM by several orders of magnitude. Yet,
these methods face two primary limitations: (1) the parameter count is
lower-bounded by the rank one decomposition, and (2) the extent of reduction is
heavily influenced by both the model architecture and the chosen rank. We
introduce NOLA, which overcomes the rank one lower bound present in LoRA. It
achieves this by re-parameterizing the low-rank matrices in LoRA using linear
combinations of randomly generated matrices (basis) and optimizing the linear
mixture coefficients only. This approach allows us to decouple the number of
trainable parameters from both the choice of rank and the network architecture.
We present adaptation results using GPT-2, LLaMA-2, and ViT in natural language
and computer vision tasks. NOLA performs as well as LoRA models with much fewer
number of parameters compared to LoRA with rank one, the best compression LoRA
can archive. Particularly, on LLaMA-2 70B, our method is almost 20 times more
compact than the most compressed LoRA without degradation in accuracy. Our code
is available here: https://github.com/UCDvision/NOLA | cs.CL | NOLA: Compressing LoRA using Linear Combination of Random Basis |
2023-10-04T01:58:34Z | http://arxiv.org/pdf/2310.02527v1 | http://arxiv.org/abs/2310.02527v1 | Tao Feng, Zifeng Wang, Jimeng Sun | 2023-10-04T01:58:34Z | cs.CL, cs.AI | The recent advancement of large language models (LLMs) has been achieved
through a combo of instruction tuning and human alignment. However, building
manually crafted instruction datasets and performing human alignment become the
bottleneck for scaling the development of LLMs. In this paper, we exploit the
idea of leveraging AI models in lieu of humans as the teacher to train student
LLMs. Our method is inspired by how human students refine their writing skills
by following the rubrics and learning from the revisions offered by their
tutors. Specifically, we employ a teacher LLM to create a curriculum for
instruction tuning of the student LLM, namely Curriculum Instruction TunING
(CITING). It encompasses two main steps: (1) the teacher LLM crafts the rubrics
for evaluating the answers corresponding to various types of questions, and (2)
the student LLM learns to follow the rubrics and perform self-correction from
the revision made by the teacher. We further iteratively carry out it to embody
the procedure of CITING. We compare CITING to a series of state-of-the-art
baselines on four datasets. Our method demonstrates strong improvement in terms
of articulate, in-depth, and comprehensive by GPT-4 evaluation. Specifically,
it achieves an average winning rate of 79.4% over SFT, 73.4% over RLHF, 78.1%
over RRHF, and 76.3% over RAFT, respectively. | cs.CL | CITING: Large Language Models Create Curriculum for Instruction Tuning |
2023-10-03T23:31:48Z | http://arxiv.org/pdf/2310.02489v2 | http://arxiv.org/abs/2310.02489v2 | Yiming Wang, Jinyu Li | 2024-01-06T23:41:15Z | cs.CL, cs.LG, cs.SD, eess.AS | Memory constraint of always-on devices is one of the major concerns when
deploying speech processing models on these devices. While larger models
trained with sufficiently large amount of data generally perform better, making
them fit in the device memory is a demanding challenge. In this paper, we aim
to reduce model size by reparameterizing model weights across Transformer
encoder layers and assuming a special weight composition and structure. More
specifically, inspired by ResNet and the more recent LoRA work, we propose an
approach named ResidualTransformer, where each weight matrix in a Transformer
layer comprises 1) a shared full-rank component with its adjacent layers, and
2) a unique low-rank component to itself. The low-rank matrices only account
for a small amount of model size increase. In addition, we add diagonal weight
matrices to improve modeling capacity of the low-rank matrices. Experiments of
our 10k-hour speech recognition and speech translation tasks show that the
Transformer encoder size can be reduced by ~3X with very slight performance
degradation. | cs.CL | ResidualTransformer: Residual Low-Rank Learning with Weight-Sharing for Transformer Layers |
2023-10-03T22:37:01Z | http://arxiv.org/pdf/2310.02469v1 | http://arxiv.org/abs/2310.02469v1 | Yijia Xiao, Yiqiao Jin, Yushi Bai, Yue Wu, Xianjun Yang, Xiao Luo, Wenchao Yu, Xujiang Zhao, Yanchi Liu, Haifeng Chen, Wei Wang, Wei Cheng | 2023-10-03T22:37:01Z | cs.CL, cs.AI, cs.LG | The proliferation of Large Language Models (LLMs) has driven considerable
interest in fine-tuning them with domain-specific data to create specialized
language models. Nevertheless, such domain-specific fine-tuning data often
contains sensitive personally identifiable information (PII). Direct
fine-tuning LLMs on this data without privacy protection poses a risk of
leakage. To address this challenge, we introduce Privacy Protection Language
Models (PPLM), a novel paradigm for fine-tuning LLMs that effectively injects
domain-specific knowledge while safeguarding data privacy. Our work offers a
theoretical analysis for model design and delves into various techniques such
as corpus curation, penalty-based unlikelihood in training loss, and
instruction-based tuning, etc. Extensive experiments across diverse datasets
and scenarios demonstrate the effectiveness of our approaches. In particular,
instruction tuning with both positive and negative examples, stands out as a
promising method, effectively protecting private data while enhancing the
model's knowledge. Our work underscores the potential for Large Language Models
as robust privacy protection learners. | cs.CL | Large Language Models Can Be Good Privacy Protection Learners |
2023-10-03T22:09:43Z | http://arxiv.org/pdf/2310.04452v1 | http://arxiv.org/abs/2310.04452v1 | Karina Shyrokykh, Maksym Girnyk, Lisa Dellmuth | 2023-10-03T22:09:43Z | cs.CL, cs.LG, stat.ME | To analyse large numbers of texts, social science researchers are
increasingly confronting the challenge of text classification. When manual
labeling is not possible and researchers have to find automatized ways to
classify texts, computer science provides a useful toolbox of machine-learning
methods whose performance remains understudied in the social sciences. In this
article, we compare the performance of the most widely used text classifiers by
applying them to a typical research scenario in social science research: a
relatively small labeled dataset with infrequent occurrence of categories of
interest, which is a part of a large unlabeled dataset. As an example case, we
look at Twitter communication regarding climate change, a topic of increasing
scholarly interest in interdisciplinary social science research. Using a novel
dataset including 5,750 tweets from various international organizations
regarding the highly ambiguous concept of climate change, we evaluate the
performance of methods in automatically classifying tweets based on whether
they are about climate change or not. In this context, we highlight two main
findings. First, supervised machine-learning methods perform better than
state-of-the-art lexicons, in particular as class balance increases. Second,
traditional machine-learning methods, such as logistic regression and random
forest, perform similarly to sophisticated deep-learning methods, whilst
requiring much less training time and computational resources. The results have
important implications for the analysis of short texts in social science
research. | cs.CL | Short text classification with machine learning in the social sciences: The case of climate change on Twitter |
2023-10-03T22:02:17Z | http://arxiv.org/pdf/2310.02457v2 | http://arxiv.org/abs/2310.02457v2 | Hannah Rose Kirk, Bertie Vidgen, Paul Röttger, Scott A. Hale | 2023-11-15T18:02:03Z | cs.CL, cs.CY | In this paper, we address the concept of "alignment" in large language models
(LLMs) through the lens of post-structuralist socio-political theory,
specifically examining its parallels to empty signifiers. To establish a shared
vocabulary around how abstract concepts of alignment are operationalised in
empirical datasets, we propose a framework that demarcates: 1) which dimensions
of model behaviour are considered important, then 2) how meanings and
definitions are ascribed to these dimensions, and by whom. We situate existing
empirical literature and provide guidance on deciding which paradigm to follow.
Through this framework, we aim to foster a culture of transparency and critical
evaluation, aiding the community in navigating the complexities of aligning
LLMs with human populations. | cs.CL | The Empty Signifier Problem: Towards Clearer Paradigms for Operationalising "Alignment" in Large Language Models |
2023-10-03T21:40:44Z | http://arxiv.org/pdf/2310.02451v1 | http://arxiv.org/abs/2310.02451v1 | Xiruo Ding, Zhecheng Sheng, Meliha Yetişgen, Serguei Pakhomov, Trevor Cohen | 2023-10-03T21:40:44Z | cs.CL | Natural Language Processing (NLP) methods have been broadly applied to
clinical tasks. Machine learning and deep learning approaches have been used to
improve the performance of clinical NLP. However, these approaches require
sufficiently large datasets for training, and trained models have been shown to
transfer poorly across sites. These issues have led to the promotion of data
collection and integration across different institutions for accurate and
portable models. However, this can introduce a form of bias called confounding
by provenance. When source-specific data distributions differ at deployment,
this may harm model performance. To address this issue, we evaluate the utility
of backdoor adjustment for text classification in a multi-site dataset of
clinical notes annotated for mentions of substance abuse. Using an evaluation
framework devised to measure robustness to distributional shifts, we assess the
utility of backdoor adjustment. Our results indicate that backdoor adjustment
can effectively mitigate for confounding shift. | cs.CL | Backdoor Adjustment of Confounding by Provenance for Robust Text Classification of Multi-institutional Clinical Notes |
2023-10-03T21:30:56Z | http://arxiv.org/pdf/2310.02446v2 | http://arxiv.org/abs/2310.02446v2 | Zheng-Xin Yong, Cristina Menghini, Stephen H. Bach | 2024-01-27T22:54:52Z | cs.CL, cs.AI, cs.CR, cs.LG | AI safety training and red-teaming of large language models (LLMs) are
measures to mitigate the generation of unsafe content. Our work exposes the
inherent cross-lingual vulnerability of these safety mechanisms, resulting from
the linguistic inequality of safety training data, by successfully
circumventing GPT-4's safeguard through translating unsafe English inputs into
low-resource languages. On the AdvBenchmark, GPT-4 engages with the unsafe
translated inputs and provides actionable items that can get the users towards
their harmful goals 79% of the time, which is on par with or even surpassing
state-of-the-art jailbreaking attacks. Other high-/mid-resource languages have
significantly lower attack success rate, which suggests that the cross-lingual
vulnerability mainly applies to low-resource languages. Previously, limited
training on low-resource languages primarily affects speakers of those
languages, causing technological disparities. However, our work highlights a
crucial shift: this deficiency now poses a risk to all LLMs users. Publicly
available translation APIs enable anyone to exploit LLMs' safety
vulnerabilities. Therefore, our work calls for a more holistic red-teaming
efforts to develop robust multilingual safeguards with wide language coverage. | cs.CL | Low-Resource Languages Jailbreak GPT-4 |
2023-10-03T21:19:50Z | http://arxiv.org/pdf/2310.02439v1 | http://arxiv.org/abs/2310.02439v1 | Naiming Liu, Shashank Sonkar, Zichao Wang, Simon Woodhead, Richard G. Baraniuk | 2023-10-03T21:19:50Z | cs.CL | We propose novel evaluations for mathematical reasoning capabilities of Large
Language Models (LLMs) based on mathematical misconceptions. Our primary
approach is to simulate LLMs as a novice learner and an expert tutor, aiming to
identify the incorrect answer to math question resulted from a specific
misconception and to recognize the misconception(s) behind an incorrect answer,
respectively. Contrary to traditional LLMs-based mathematical evaluations that
focus on answering math questions correctly, our approach takes inspirations
from principles in educational learning sciences. We explicitly ask LLMs to
mimic a novice learner by answering questions in a specific incorrect manner
based on incomplete knowledge; and to mimic an expert tutor by identifying
misconception(s) corresponding to an incorrect answer to a question. Using
simple grade-school math problems, our experiments reveal that, while LLMs can
easily answer these questions correctly, they struggle to identify 1) the
incorrect answer corresponding to specific incomplete knowledge
(misconceptions); 2) the misconceptions that explain particular incorrect
answers. Our study indicates new opportunities for enhancing LLMs' math
reasoning capabilities, especially on developing robust student simulation and
expert tutoring models in the educational applications such as intelligent
tutoring systems. | cs.CL | Novice Learner and Expert Tutor: Evaluating Math Reasoning Abilities of Large Language Models with Misconceptions |
2023-10-03T20:54:29Z | http://arxiv.org/pdf/2310.02431v1 | http://arxiv.org/abs/2310.02431v1 | Yufan Chen, Arjun Arunasalam, Z. Berkay Celik | 2023-10-03T20:54:29Z | cs.HC, cs.CL | Users seek security & privacy (S&P) advice from online resources, including
trusted websites and content-sharing platforms. These resources help users
understand S&P technologies and tools and suggest actionable strategies. Large
Language Models (LLMs) have recently emerged as trusted information sources.
However, their accuracy and correctness have been called into question. Prior
research has outlined the shortcomings of LLMs in answering multiple-choice
questions and user ability to inadvertently circumvent model restrictions
(e.g., to produce toxic content). Yet, the ability of LLMs to provide reliable
S&P advice is not well-explored. In this paper, we measure their ability to
refute popular S&P misconceptions that the general public holds. We first study
recent academic literature to curate a dataset of over a hundred S&P-related
misconceptions across six different topics. We then query two popular LLMs
(Bard and ChatGPT) and develop a labeling guide to evaluate their responses to
these misconceptions. To comprehensively evaluate their responses, we further
apply three strategies: query each misconception multiple times, generate and
query their paraphrases, and solicit source URLs of the responses. Both models
demonstrate, on average, a 21.3% non-negligible error rate, incorrectly
supporting popular S&P misconceptions. The error rate increases to 32.6% when
we repeatedly query LLMs with the same or paraphrased misconceptions. We also
expose that models may partially support a misconception or remain
noncommittal, refusing a firm stance on misconceptions. Our exploration of
information sources for responses revealed that LLMs are susceptible to
providing invalid URLs (21.2% for Bard and 67.7% for ChatGPT) or point to
unrelated sources (44.2% returned by Bard and 18.3% by ChatGPT). | cs.HC | Can Large Language Models Provide Security & Privacy Advice? Measuring the Ability of LLMs to Refute Misconceptions |
2023-10-03T20:34:59Z | http://arxiv.org/pdf/2310.02421v1 | http://arxiv.org/abs/2310.02421v1 | Sia Gholami, Marwan Omar | 2023-10-03T20:34:59Z | cs.LG, cs.AI, cs.CL | The burgeoning complexity of contemporary deep learning models, while
achieving unparalleled accuracy, has inadvertently introduced deployment
challenges in resource-constrained environments. Knowledge distillation, a
technique aiming to transfer knowledge from a high-capacity "teacher" model to
a streamlined "student" model, emerges as a promising solution to this dilemma.
This paper provides a comprehensive overview of the knowledge distillation
paradigm, emphasizing its foundational principles such as the utility of soft
labels and the significance of temperature scaling. Through meticulous
examination, we elucidate the critical determinants of successful distillation,
including the architecture of the student model, the caliber of the teacher,
and the delicate balance of hyperparameters. While acknowledging its profound
advantages, we also delve into the complexities and challenges inherent in the
process. Our exploration underscores knowledge distillation's potential as a
pivotal technique in optimizing the trade-off between model performance and
deployment efficiency. | cs.LG | Can a student Large Language Model perform as well as it's teacher? |
2023-10-03T20:11:23Z | http://arxiv.org/pdf/2310.02410v1 | http://arxiv.org/abs/2310.02410v1 | Young Jin Kim, Raffy Fahim, Hany Hassan Awadalla | 2023-10-03T20:11:23Z | cs.LG, cs.CL | Large Mixture of Experts (MoE) models could achieve state-of-the-art quality
on various language tasks, including machine translation task, thanks to the
efficient model scaling capability with expert parallelism. However, it has
brought a fundamental issue of larger memory consumption and increased memory
bandwidth bottleneck at deployment time. In this paper, we propose Mixture of
Quantized Experts (MoQE) which is a simple weight-only quantization method
applying ultra low-bit down to 2-bit quantizations only to expert weights for
mitigating the increased memory and latency issues of MoE models. We show that
low-bit quantization together with the MoE architecture delivers a reliable
model performance while reducing the memory size significantly even without any
additional training in most cases. In particular, expert layers in MoE models
are much more robust to the quantization than conventional feedforward networks
(FFN) layers. In our comprehensive analysis, we show that MoE models with 2-bit
expert weights can deliver better model performance than the dense model
trained on the same dataset. As a result of low-bit quantization, we show the
model size can be reduced by 79.6% of the original half precision floating
point (fp16) MoE model. Combined with an optimized GPU runtime implementation,
it also achieves 1.24X speed-up on A100 GPUs. | cs.LG | Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness |
2023-10-03T20:07:06Z | http://arxiv.org/pdf/2310.02409v2 | http://arxiv.org/abs/2310.02409v2 | Guanghui Qin, Corby Rosset, Ethan C. Chau, Nikhil Rao, Benjamin Van Durme | 2024-06-13T15:19:24Z | cs.CL, cs.AI, cs.LG, I.2.7; I.2.6 | Transformer-based language models (LMs) are inefficient in long contexts. We
propose Dodo, a solution for context compression. Instead of one vector per
token in a standard transformer model, Dodo represents text with a dynamic
number of hidden states at each layer, reducing the cost of self-attention to a
fraction of typical time and space. Moreover, off-the-shelf models such as
LLaMA can be adapted to Dodo by efficient parameter tuning methods such as
LoRA. In use, Dodo can act as either an autoregressive LM or a context
compressor for downstream tasks. We demonstrate through experiments in language
modeling, question answering, and summarization that Dodo retains capabilities
in these tasks, while drastically reducing the overhead during decoding. For
example, in the autoencoding task, Dodo shrinks context at a 20x compression
ratio with a BLEU score of 98% for reconstruction, achieving nearly lossless
encoding. | cs.CL | Dodo: Dynamic Contextual Compression for Decoder-only LMs |
2023-10-03T20:03:08Z | http://arxiv.org/pdf/2310.02408v1 | http://arxiv.org/abs/2310.02408v1 | Giacomo Ibba, Sabrina Aufiero, Silvia Bartolucci, Rumyana Neykova, Marco Ortu, Roberto Tonelli, Giuseppe Destefanis | 2023-10-03T20:03:08Z | cs.IT, cs.CL, math.IT | This paper presents MindTheDApp, a toolchain designed specifically for the
structural analysis of Ethereum-based Decentralized Applications (DApps), with
a distinct focus on a complex network-driven approach. Unlike existing tools,
our toolchain combines the power of ANTLR4 and Abstract Syntax Tree (AST)
traversal techniques to transform the architecture and interactions within
smart contracts into a specialized bipartite graph. This enables advanced
network analytics to highlight operational efficiencies within the DApp's
architecture.
The bipartite graph generated by the proposed tool comprises two sets of
nodes: one representing smart contracts, interfaces, and libraries, and the
other including functions, events, and modifiers. Edges in the graph connect
functions to smart contracts they interact with, offering a granular view of
interdependencies and execution flow within the DApp. This network-centric
approach allows researchers and practitioners to apply complex network theory
in understanding the robustness, adaptability, and intricacies of decentralized
systems.
Our work contributes to the enhancement of security in smart contracts by
allowing the visualisation of the network, and it provides a deep understanding
of the architecture and operational logic within DApps. Given the growing
importance of smart contracts in the blockchain ecosystem and the emerging
application of complex network theory in technology, our toolchain offers a
timely contribution to both academic research and practical applications in the
field of blockchain technology. | cs.IT | MindTheDApp: A Toolchain for Complex Network-Driven Structural Analysis of Ethereum-based Decentralised Applications |
2023-10-03T19:44:37Z | http://arxiv.org/pdf/2310.04451v2 | http://arxiv.org/abs/2310.04451v2 | Xiaogeng Liu, Nan Xu, Muhao Chen, Chaowei Xiao | 2024-03-20T21:34:56Z | cs.CL, cs.AI | The aligned Large Language Models (LLMs) are powerful language understanding
and decision-making tools that are created through extensive alignment with
human feedback. However, these large models remain susceptible to jailbreak
attacks, where adversaries manipulate prompts to elicit malicious outputs that
should not be given by aligned LLMs. Investigating jailbreak prompts can lead
us to delve into the limitations of LLMs and further guide us to secure them.
Unfortunately, existing jailbreak techniques suffer from either (1) scalability
issues, where attacks heavily rely on manual crafting of prompts, or (2)
stealthiness problems, as attacks depend on token-based algorithms to generate
prompts that are often semantically meaningless, making them susceptible to
detection through basic perplexity testing. In light of these challenges, we
intend to answer this question: Can we develop an approach that can
automatically generate stealthy jailbreak prompts? In this paper, we introduce
AutoDAN, a novel jailbreak attack against aligned LLMs. AutoDAN can
automatically generate stealthy jailbreak prompts by the carefully designed
hierarchical genetic algorithm. Extensive evaluations demonstrate that AutoDAN
not only automates the process while preserving semantic meaningfulness, but
also demonstrates superior attack strength in cross-model transferability, and
cross-sample universality compared with the baseline. Moreover, we also compare
AutoDAN with perplexity-based defense methods and show that AutoDAN can bypass
them effectively. | cs.CL | AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models |
2023-10-03T19:05:32Z | http://arxiv.org/pdf/2310.02382v1 | http://arxiv.org/abs/2310.02382v1 | Liming Wang, Mark Hasegawa-Johnson, Chang D. Yoo | 2023-10-03T19:05:32Z | cs.CL, cs.SD, eess.AS | Training unsupervised speech recognition systems presents challenges due to
GAN-associated instability, misalignment between speech and text, and
significant memory demands. To tackle these challenges, we introduce a novel
ASR system, ESPUM. This system harnesses the power of lower-order N-skipgrams
(up to N=3) combined with positional unigram statistics gathered from a small
batch of samples. Evaluated on the TIMIT benchmark, our model showcases
competitive performance in ASR and phoneme segmentation tasks. Access our
publicly available code at https://github.com/lwang114/GraphUnsupASR. | cs.CL | Unsupervised Speech Recognition with N-Skipgram and Positional Unigram Matching |
2023-10-03T18:54:10Z | http://arxiv.org/pdf/2310.02374v4 | http://arxiv.org/abs/2310.02374v4 | Mahyar Abbasian, Iman Azimi, Amir M. Rahmani, Ramesh Jain | 2024-01-23T21:27:14Z | cs.CL | Conversational Health Agents (CHAs) are interactive systems that provide
healthcare services, such as assistance and diagnosis. Current CHAs, especially
those utilizing Large Language Models (LLMs), primarily focus on conversation
aspects. However, they offer limited agent capabilities, specifically lacking
multi-step problem-solving, personalized conversations, and multimodal data
analysis. Our aim is to overcome these limitations. We propose openCHA, an
open-source LLM-powered framework, to empower conversational agents to generate
a personalized response for users' healthcare queries. This framework enables
developers to integrate external sources including data sources, knowledge
bases, and analysis models, into their LLM-based solutions. openCHA includes an
orchestrator to plan and execute actions for gathering information from
external sources, essential for formulating responses to user inquiries. It
facilitates knowledge acquisition, problem-solving capabilities, multilingual
and multimodal conversations, and fosters interaction with various AI
platforms. We illustrate the framework's proficiency in handling complex
healthcare tasks via three demonstrations. Moreover, we release openCHA as open
source available to the community via GitHub. | cs.CL | Conversational Health Agents: A Personalized LLM-Powered Agent Framework |
2023-10-03T18:52:19Z | http://arxiv.org/pdf/2310.02372v1 | http://arxiv.org/abs/2310.02372v1 | Ritesh Kumar, Saurabh Goyal, Ashish Verma, Vatche Isahagian | 2023-10-03T18:52:19Z | cs.CL, cs.AI | Key value pair (KVP) extraction or Named Entity Recognition(NER) from
visually rich documents has been an active area of research in document
understanding and data extraction domain. Several transformer based models such
as LayoutLMv2, LayoutLMv3, and LiLT have emerged achieving state of the art
results. However, addition of even a single new class to the existing model
requires (a) re-annotation of entire training dataset to include this new class
and (b) retraining the model again. Both of these issues really slow down the
deployment of updated model. \\ We present \textbf{ProtoNER}: Prototypical
Network based end-to-end KVP extraction model that allows addition of new
classes to an existing model while requiring minimal number of newly annotated
training samples. The key contributions of our model are: (1) No dependency on
dataset used for initial training of the model, which alleviates the need to
retain original training dataset for longer duration as well as data
re-annotation which is very time consuming task, (2) No intermediate synthetic
data generation which tends to add noise and results in model's performance
degradation, and (3) Hybrid loss function which allows model to retain
knowledge about older classes as well as learn about newly added classes.\\
Experimental results show that ProtoNER finetuned with just 30 samples is able
to achieve similar results for the newly added classes as that of regular model
finetuned with 2600 samples. | cs.CL | ProtoNER: Few shot Incremental Learning for Named Entity Recognition using Prototypical Networks |
2023-10-03T18:32:34Z | http://arxiv.org/pdf/2310.02357v3 | http://arxiv.org/abs/2310.02357v3 | Sergey Berezin, Reza Farahbakhsh, Noel Crespi | 2023-10-19T18:38:09Z | cs.CL, cs.AI, cs.LG | The fundamental problem in toxicity detection task lies in the fact that the
toxicity is ill-defined. This causes us to rely on subjective and vague data in
models' training, which results in non-robust and non-accurate results: garbage
in - garbage out.
This work suggests a new, stress-level-based definition of toxicity designed
to be objective and context-aware. On par with it, we also describe possible
ways of applying this new definition to dataset creation and model training. | cs.CL | On the definition of toxicity in NLP |
2023-10-03T17:59:46Z | http://arxiv.org/pdf/2310.02263v2 | http://arxiv.org/abs/2310.02263v2 | Canwen Xu, Corby Rosset, Ethan C. Chau, Luciano Del Corro, Shweti Mahajan, Julian McAuley, Jennifer Neville, Ahmed Hassan Awadallah, Nikhil Rao | 2024-04-03T00:16:19Z | cs.CL, cs.AI, cs.LG | Alignment serves as an important step to steer large language models (LLMs)
towards human preferences. In this paper, we propose an automatic way to
construct contrastive data for LLM, using preference pairs from multiple models
of varying strengths (e.g., InstructGPT, ChatGPT and GPT-4). We compare the
contrastive techniques of SLiC and DPO to SFT baselines and find that DPO
provides a step-function improvement even after continuing SFT saturates. We
also explore a data curriculum learning scheme for contrastive post-training,
which starts by learning from "easier" pairs and transitioning to "harder"
ones, which further improves alignment. Finally, we scale up our experiments to
train with more data and larger models like Orca. Remarkably, our automatic
contrastive post-training further improves the performance of Orca, already a
state-of-the-art instruction learning model tuned with GPT-4 outputs, to
outperform ChatGPT. | cs.CL | Automatic Pair Construction for Contrastive Post-training |
2023-10-03T17:59:46Z | http://arxiv.org/pdf/2310.02264v1 | http://arxiv.org/abs/2310.02264v1 | Haoyu Zhou, Mingyu Ding, Weikun Peng, Masayoshi Tomizuka, Lin Shao, Chuang Gan | 2023-10-03T17:59:46Z | cs.RO, cs.CL, cs.CV, cs.LG | This work introduces a framework harnessing the capabilities of Large
Language Models (LLMs) to generate primitive task conditions for generalizable
long-horizon manipulations with novel objects and unseen tasks. These task
conditions serve as guides for the generation and adjustment of Dynamic
Movement Primitives (DMP) trajectories for long-horizon task execution. We
further create a challenging robotic manipulation task suite based on Pybullet
for long-horizon task evaluation. Extensive experiments in both simulated and
real-world environments demonstrate the effectiveness of our framework on both
familiar tasks involving new objects and novel but related tasks, highlighting
the potential of LLMs in enhancing robotic system versatility and adaptability.
Project website: https://object814.github.io/Task-Condition-With-LLM/ | cs.RO | Generalizable Long-Horizon Manipulations with Large Language Models |
2023-10-03T17:59:32Z | http://arxiv.org/pdf/2310.02304v2 | http://arxiv.org/abs/2310.02304v2 | Eric Zelikman, Eliana Lorch, Lester Mackey, Adam Tauman Kalai | 2024-03-01T17:11:21Z | cs.CL, cs.AI, cs.LG, stat.ML | Several recent advances in AI systems (e.g., Tree-of-Thoughts and
Program-Aided Language Models) solve problems by providing a "scaffolding"
program that structures multiple calls to language models to generate better
outputs. A scaffolding program is written in a programming language such as
Python. In this work, we use a language-model-infused scaffolding program to
improve itself. We start with a seed "improver" that improves an input program
according to a given utility function by querying a language model several
times and returning the best solution. We then run this seed improver to
improve itself. Across a small set of downstream tasks, the resulting improved
improver generates programs with significantly better performance than its seed
improver. A variety of self-improvement strategies are proposed by the language
model, including beam search, genetic algorithms, and simulated annealing.
Since the language models themselves are not altered, this is not full
recursive self-improvement. Nonetheless, it demonstrates that a modern language
model, GPT-4 in our experiments, is capable of writing code that can call
itself to improve itself. We consider concerns around the development of
self-improving technologies and evaluate the frequency with which the generated
code bypasses a sandbox. | cs.CL | Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.