PyTorchKR Papers

최신 논문

총 11,113편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2402.16906공식 코드 · Papers with Code

Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step

Large language models (LLMs) are leading significant progress in code generation. Beyond one-pass code generation, recent works further integrate unit tests and program verifiers into LLMs to iteratively refine the generated programs. However, these works consider the generated programs as an indivisible entity, which falls short for LLMs in debugging the programs, especially when the programs contain complex logic flows and data operations. In contrast, when human developers debug programs, they typically set breakpoints and selectively examine runtime execution information. The execution flow and the intermediate variables play a crucial role in the debugging process, yet they are underutilized in the existing literature on code generation. In this study, we introduce Large Language Model Debugger (LDB), a novel debugging framework that enables LLMs to refine their generated programs with the runtime execution information. Specifically, LDB segments the programs into basic blocks and tracks the values of intermediate variables after each block throughout the runtime execution. This allows LLMs to concentrate on simpler code units within the overall execution flow, verify their correctness against the task description block by block, and efficiently pinpoint any potential errors. Experiments demonstrate that LDB consistently enhances the baseline performance by up to 9.8% across the HumanEval, MBPP, and TransCoder benchmarks, archiving new state-of-the-art performance in code debugging for various LLM selections.

arXiv:2402.16832공식 코드 · Papers with Code

Cross-Modal Projection in Multimodal LLMs Doesn't Really Project Visual Attributes to Textual Space

Multimodal large language models (MLLMs) like LLaVA and GPT-4(V) enable general-purpose conversations about images with the language modality. As off-the-shelf MLLMs may have limited capabilities on images from domains like dermatology and agriculture, they must be fine-tuned to unlock domain-specific applications. The prevalent architecture of current open-source MLLMs comprises two major modules: an image-language (cross-modal) projection network and a large language model. It is desirable to understand the roles of these two modules in modeling domain-specific visual attributes to inform the design of future models and streamline the interpretability efforts on the current models. To this end, via experiments on 4 datasets and under 2 fine-tuning settings, we find that as the MLLM is fine-tuned, it indeed gains domain-specific visual capabilities, but the updates do not lead to the projection extracting relevant domain-specific visual attributes. Our results indicate that the domain-specific visual attributes are modeled by the LLM, even when only the projection is fine-tuned. Through this study, we offer a potential reinterpretation of the role of cross-modal projections in MLLM architectures. Project webpage: https://claws-lab.github.io/projection-in-MLLMs/

arXiv:2402.16667공식 코드 · Papers with Code

RepoAgent: An LLM-Powered Open-Source Framework for Repository-level Code Documentation Generation

Generative models have demonstrated considerable potential in software engineering, particularly in tasks such as code generation and debugging. However, their utilization in the domain of code documentation generation remains underexplored. To this end, we introduce RepoAgent, a large language model powered open-source framework aimed at proactively generating, maintaining, and updating code documentation. Through both qualitative and quantitative evaluations, we have validated the effectiveness of our approach, showing that RepoAgent excels in generating high-quality repository-level documentation. The code and results are publicly accessible at https://github.com/OpenBMB/RepoAgent.

arXiv:2402.17645공식 코드 · Papers with Code

SongComposer: A Large Language Model for Lyric and Melody Generation in Song Composition

Creating lyrics and melodies for the vocal track in a symbolic format, known as song composition, demands expert musical knowledge of melody, an advanced understanding of lyrics, and precise alignment between them. Despite achievements in sub-tasks such as lyric generation, lyric-to-melody, and melody-to-lyric, etc, a unified model for song composition has not yet been achieved. In this paper, we introduce SongComposer, a pioneering step towards a unified song composition model that can readily create symbolic lyrics and melodies following instructions. SongComposer is a music-specialized large language model (LLM) that, for the first time, integrates the capability of simultaneously composing lyrics and melodies into LLMs by leveraging three key innovations: 1) a flexible tuple format for word-level alignment of lyrics and melodies, 2) an extended tokenizer vocabulary for song notes, with scalar initialization based on musical knowledge to capture rhythm, and 3) a multi-stage pipeline that captures musical structure, starting with motif-level melody patterns and progressing to phrase-level structure for improved coherence. Extensive experiments demonstrate that SongComposer outperforms advanced LLMs, including GPT-4, in tasks such as lyric-to-melody generation, melody-to-lyric generation, song continuation, and text-to-song creation. Moreover, we will release SongCompose, a large-scale dataset for training, containing paired lyrics and melodies in Chinese and English.

arXiv:2402.17407

A Neural Rewriting System to Solve Algorithmic Problems

Modern neural network architectures still struggle to learn algorithmic procedures that require to systematically apply compositional rules to solve out-of-distribution problem instances. In this work, we focus on formula simplification problems, a class of synthetic benchmarks used to study the systematic generalization capabilities of neural architectures. We propose a modular architecture designed to learn a general procedure for solving nested mathematical formulas by only relying on a minimal set of training examples. Inspired by rewriting systems, a classic framework in symbolic artificial intelligence, we include in the architecture three specialized and interacting modules: the Selector, trained to identify solvable sub-expressions; the Solver, mapping sub-expressions to their values; and the Combiner, replacing sub-expressions in the original formula with the solution provided by the Solver. We benchmark our system against the Neural Data Router, a recent model specialized for systematic generalization, and a state-of-the-art large language model (GPT-4) probed with advanced prompting strategies. We demonstrate that our approach achieves a higher degree of out-of-distribution generalization compared to these alternative approaches on three different types of formula simplification problems, and we discuss its limitations by analyzing its failures.

arXiv:2402.17161

Large Language Model for Participatory Urban Planning

Participatory urban planning is the mainstream of modern urban planning that involves the active engagement of residents. However, the traditional participatory paradigm requires experienced planning experts and is often time-consuming and costly. Fortunately, the emerging Large Language Models (LLMs) have shown considerable ability to simulate human-like agents, which can be used to emulate the participatory process easily. In this work, we introduce an LLM-based multi-agent collaboration framework for participatory urban planning, which can generate land-use plans for urban regions considering the diverse needs of residents. Specifically, we construct LLM agents to simulate a planner and thousands of residents with diverse profiles and backgrounds. We first ask the planner to carry out an initial land-use plan. To deal with the different facilities needs of residents, we initiate a discussion among the residents in each community about the plan, where residents provide feedback based on their profiles. Furthermore, to improve the efficiency of discussion, we adopt a fishbowl discussion mechanism, where part of the residents discuss and the rest of them act as listeners in each round. Finally, we let the planner modify the plan based on residents' feedback. We deploy our method on two real-world regions in Beijing. Experiments show that our method achieves state-of-the-art performance in residents satisfaction and inclusion metrics, and also outperforms human experts in terms of service accessibility and ecology metrics.

arXiv:2402.16006

ASETF: A Novel Method for Jailbreak Attack on LLMs through Translate Suffix Embeddings

The safety defense methods of Large language models(LLMs) stays limited because the dangerous prompts are manually curated to just few known attack types, which fails to keep pace with emerging varieties. Recent studies found that attaching suffixes to harmful instructions can hack the defense of LLMs and lead to dangerous outputs. However, similar to traditional text adversarial attacks, this approach, while effective, is limited by the challenge of the discrete tokens. This gradient based discrete optimization attack requires over 100,000 LLM calls, and due to the unreadable of adversarial suffixes, it can be relatively easily penetrated by common defense methods such as perplexity filters. To cope with this challenge, in this paper, we proposes an Adversarial Suffix Embedding Translation Framework (ASETF), aimed at transforming continuous adversarial suffix embeddings into coherent and understandable text. This method greatly reduces the computational overhead during the attack process and helps to automatically generate multiple adversarial samples, which can be used as data to strengthen LLMs security defense. Experimental evaluations were conducted on Llama2, Vicuna, and other prominent LLMs, employing harmful directives sourced from the Advbench dataset. The results indicate that our method significantly reduces the computation time of adversarial suffixes and achieves a much better attack success rate to existing techniques, while significantly enhancing the textual fluency of the prompts. In addition, our approach can be generalized into a broader method for generating transferable adversarial suffixes that can successfully attack multiple LLMs, even black-box LLMs, such as ChatGPT and Gemini.

arXiv:2402.15911

PRP: Propagating Universal Perturbations to Attack Large Language Model Guard-Rails

Large language models (LLMs) are typically aligned to be harmless to humans. Unfortunately, recent work has shown that such models are susceptible to automated jailbreak attacks that induce them to generate harmful content. More recent LLMs often incorporate an additional layer of defense, a Guard Model, which is a second LLM that is designed to check and moderate the output response of the primary LLM. Our key contribution is to show a novel attack strategy, PRP, that is successful against several open-source (e.g., Llama 2) and closed-source (e.g., GPT 3.5) implementations of Guard Models. PRP leverages a two step prefix-based attack that operates by (a) constructing a universal adversarial prefix for the Guard Model, and (b) propagating this prefix to the response. We find that this procedure is effective across multiple threat models, including ones in which the adversary has no access to the Guard Model at all. Our work suggests that further advances are required on defenses and Guard Models before they can be considered effective.

arXiv:2402.15754

HD-Eval: Aligning Large Language Model Evaluators Through Hierarchical Criteria Decomposition

Large language models (LLMs) have emerged as a promising alternative to expensive human evaluations. However, the alignment and coverage of LLM-based evaluations are often limited by the scope and potential bias of the evaluation prompts and criteria. To address this challenge, we propose HD-Eval, a novel framework that iteratively aligns LLM-based evaluators with human preference via Hierarchical Criteria Decomposition. HD-Eval inherits the essence from the evaluation mindset of human experts and enhances the alignment of LLM-based evaluators by decomposing a given evaluation task into finer-grained criteria, aggregating them according to estimated human preferences, pruning insignificant criteria with attribution, and further decomposing significant criteria. By integrating these steps within an iterative alignment training process, we obtain a hierarchical decomposition of criteria that comprehensively captures aspects of natural language at multiple levels of granularity. Implemented as a white box, the human preference-guided aggregator is efficient to train and more explainable than relying solely on prompting, and its independence from model parameters makes it applicable to closed-source LLMs. Extensive experiments on three evaluation domains demonstrate the superiority of HD-Eval in further aligning state-of-the-art evaluators and providing deeper insights into the explanation of evaluation results and the task itself.

arXiv:2402.15627공식 코드 · Papers with Code

MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs

We present the design, implementation and engineering experience in building and deploying MegaScale, a production system for training large language models (LLMs) at the scale of more than 10,000 GPUs. Training LLMs at this scale brings unprecedented challenges to training efficiency and stability. We take a full-stack approach that co-designs the algorithmic and system components across model block and optimizer design, computation and communication overlapping, operator optimization, data pipeline, and network performance tuning. Maintaining high efficiency throughout the training process (i.e., stability) is an important consideration in production given the long extent of LLM training jobs. Many hard stability issues only emerge at large scale, and in-depth observability is the key to address them. We develop a set of diagnosis tools to monitor system components and events deep in the stack, identify root causes, and derive effective techniques to achieve fault tolerance and mitigate stragglers. MegaScale achieves 55.2% Model FLOPs Utilization (MFU) when training a 175B LLM model on 12,288 GPUs, improving the MFU by 1.34x compared to Megatron-LM. We share our operational experience in identifying and fixing failures and stragglers. We hope by articulating the problems and sharing our experience from a systems perspective, this work can inspire future LLM systems research.

arXiv:2402.15809공식 코드 · Papers with Code

Empowering Large Language Model Agents through Action Learning

Large Language Model (LLM) Agents have recently garnered increasing interest yet they are limited in their ability to learn from trial and error, a key element of intelligent behavior. In this work, we argue that the capacity to learn new actions from experience is fundamental to the advancement of learning in LLM agents. While humans naturally expand their action spaces and develop skills through experiential learning, LLM agents typically operate within fixed action spaces, limiting their potential for growth. To address these challenges, our study explores open-action learning for language agents. We introduce a framework LearnAct with an iterative learning strategy to create and improve actions in the form of Python functions. In each iteration, LLM revises and updates the currently available actions based on the errors identified in unsuccessful training tasks, thereby enhancing action effectiveness. Our experimental evaluations across Robotic Planning and Alfworld environments reveal that after learning on a few training task instances, our approach to open-action learning markedly improves agent performance for the type of task (by 32 percent in AlfWorld compared to ReAct+Reflexion, for instance) highlighting the importance of experiential action learning in the development of more intelligent LLM agents.

arXiv:2402.15759공식 코드 · Papers with Code

TV-SAM: Increasing Zero-Shot Segmentation Performance on Multimodal Medical Images Using GPT-4 Generated Descriptive Prompts Without Human Annotation

This study presents a novel multimodal medical image zero-shot segmentation algorithm named the text-visual-prompt segment anything model (TV-SAM) without any manual annotations. The TV-SAM incorporates and integrates the large language model GPT-4, the vision language model GLIP, and the SAM to autonomously generate descriptive text prompts and visual bounding box prompts from medical images, thereby enhancing the SAM's capability for zero-shot segmentation. Comprehensive evaluations are implemented on seven public datasets encompassing eight imaging modalities to demonstrate that TV-SAM can effectively segment unseen targets across various modalities without additional training. TV-SAM significantly outperforms SAM AUTO and GSAM, closely matching the performance of SAM BBOX with gold standard bounding box prompts and surpasses the state-of-the-art methods on specific datasets such as ISIC and WBC. The study indicates that TV-SAM serves as an effective multimodal medical image zero-shot segmentation algorithm, highlighting the significant contribution of GPT-4 to zero-shot segmentation. By integrating foundational models such as GPT-4, GLIP, and SAM, the ability to address complex problems in specialized domains can be enhanced.

arXiv:2402.15690

Foot In The Door: Understanding Large Language Model Jailbreaking via Cognitive Psychology

Large Language Models (LLMs) have gradually become the gateway for people to acquire new knowledge. However, attackers can break the model's security protection ("jail") to access restricted information, which is called "jailbreaking." Previous studies have shown the weakness of current LLMs when confronted with such jailbreaking attacks. Nevertheless, comprehension of the intrinsic decision-making mechanism within the LLMs upon receipt of jailbreak prompts is noticeably lacking. Our research provides a psychological explanation of the jailbreak prompts. Drawing on cognitive consistency theory, we argue that the key to jailbreak is guiding the LLM to achieve cognitive coordination in an erroneous direction. Further, we propose an automatic black-box jailbreaking method based on the Foot-in-the-Door (FITD) technique. This method progressively induces the model to answer harmful questions via multi-step incremental prompts. We instantiated a prototype system to evaluate the jailbreaking effectiveness on 8 advanced LLMs, yielding an average success rate of 83.9%. This study builds a psychological perspective on the explanatory insights into the intrinsic decision-making logic of LLMs.

arXiv:2402.15631

Fine-Grained Self-Endorsement Improves Factuality and Reasoning

This work studies improving large language model (LLM) generations at inference time by mitigating fact-conflicting hallucinations. Particularly, we propose a self-endorsement framework that leverages the fine-grained fact-level comparisons across multiple sampled responses. Compared with prior ensemble methods (Wang et al., 2022;Chen et al., 2023)) that perform response-level selection, our approach can better alleviate hallucinations, especially for longform generation tasks. Our approach can broadly benefit smaller and open-source LLMs as it mainly conducts simple content-based comparisons. Experiments on Biographies show that our method can effectively improve the factuality of generations with simple and intuitive prompts across different scales of LLMs. Besides, comprehensive analyses on TriviaQA and GSM8K demonstrate the potential of self-endorsement for broader application.

arXiv:2402.16363공식 코드 · Papers with Code

LLM Inference Unveiled: Survey and Roofline Model Insights

The field of efficient Large Language Model (LLM) inference is rapidly evolving, presenting a unique blend of opportunities and challenges. Although the field has expanded and is vibrant, there hasn't been a concise framework that analyzes the various methods of LLM Inference to provide a clear understanding of this domain. Our survey stands out from traditional literature reviews by not only summarizing the current state of research but also by introducing a framework based on roofline model for systematic analysis of LLM inference techniques. This framework identifies the bottlenecks when deploying LLMs on hardware devices and provides a clear understanding of practical problems, such as why LLMs are memory-bound, how much memory and computation they need, and how to choose the right hardware. We systematically collate the latest advancements in efficient LLM inference, covering crucial areas such as model compression (e.g., Knowledge Distillation and Quantization), algorithm improvements (e.g., Early Exit and Mixture-of-Expert), and both hardware and system-level enhancements. Our survey stands out by analyzing these methods with roofline model, helping us understand their impact on memory access and computation. This distinctive approach not only showcases the current research landscape but also delivers valuable insights for practical implementation, positioning our work as an indispensable resource for researchers new to the field as well as for those seeking to deepen their understanding of efficient LLM deployment. The analyze tool, LLM-Viewer, is open-sourced.

arXiv:2402.16035

Text Understanding and Generation Using Transformer Models for Intelligent E-commerce Recommendations

With the rapid development of artificial intelligence technology, Transformer structural pre-training model has become an important tool for large language model (LLM) tasks. In the field of e-commerce, these models are especially widely used, from text understanding to generating recommendation systems, which provide powerful technical support for improving user experience and optimizing service processes. This paper reviews the core application scenarios of Transformer pre-training model in e-commerce text understanding and recommendation generation, including but not limited to automatic generation of product descriptions, sentiment analysis of user comments, construction of personalized recommendation system and automated processing of customer service conversations. Through a detailed analysis of the model's working principle, implementation process, and application effects in specific cases, this paper emphasizes the unique advantages of pre-trained models in understanding complex user intentions and improving the quality of recommendations. In addition, the challenges and improvement directions for the future are also discussed, such as how to further improve the generalization ability of the model, the ability to handle large-scale data sets, and technical strategies to protect user privacy. Ultimately, the paper points out that the application of Transformer structural pre-training models in e-commerce has not only driven technological innovation, but also brought substantial benefits to merchants and consumers, and looking forward, these models will continue to play a key role in e-commerce and beyond.

arXiv:2402.16389공식 코드 · Papers with Code

MoZIP: A Multilingual Benchmark to Evaluate Large Language Models in Intellectual Property

Large language models (LLMs) have demonstrated impressive performance in various natural language processing (NLP) tasks. However, there is limited understanding of how well LLMs perform in specific domains (e.g, the intellectual property (IP) domain). In this paper, we contribute a new benchmark, the first Multilingual-oriented quiZ on Intellectual Property (MoZIP), for the evaluation of LLMs in the IP domain. The MoZIP benchmark includes three challenging tasks: IP multiple-choice quiz (IPQuiz), IP question answering (IPQA), and patent matching (PatentMatch). In addition, we also develop a new IP-oriented multilingual large language model (called MoZi), which is a BLOOMZ-based model that has been supervised fine-tuned with multilingual IP-related text data. We evaluate our proposed MoZi model and four well-known LLMs (i.e., BLOOMZ, BELLE, ChatGLM and ChatGPT) on the MoZIP benchmark. Experimental results demonstrate that MoZi outperforms BLOOMZ, BELLE and ChatGLM by a noticeable margin, while it had lower scores compared with ChatGPT. Notably, the performance of current LLMs on the MoZIP benchmark has much room for improvement, and even the most powerful ChatGPT does not reach the passing level. Our source code, data, and models are available at \url{https://github.com/AI-for-Science/MoZi}.

arXiv:2402.14850

CHATATC: Large Language Model-Driven Conversational Agents for Supporting Strategic Air Traffic Flow Management

Generative artificial intelligence (AI) and large language models (LLMs) have gained rapid popularity through publicly available tools such as ChatGPT. The adoption of LLMs for personal and professional use is fueled by the natural interactions between human users and computer applications such as ChatGPT, along with powerful summarization and text generation capabilities. Given the widespread use of such generative AI tools, in this work we investigate how these tools can be deployed in a non-safety critical, strategic traffic flow management setting. Specifically, we train an LLM, CHATATC, based on a large historical data set of Ground Delay Program (GDP) issuances, spanning 2000-2023 and consisting of over 80,000 GDP implementations, revisions, and cancellations. We test the query and response capabilities of CHATATC, documenting successes (e.g., providing correct GDP rates, durations, and reason) and shortcomings (e.g,. superlative questions). We also detail the design of a graphical user interface for future users to interact and collaborate with the CHATATC conversational agent.

arXiv:2402.15215공식 코드 · Papers with Code

Item-side Fairness of Large Language Model-based Recommendation System

Recommendation systems for Web content distribution intricately connect to the information access and exposure opportunities for vulnerable populations. The emergence of Large Language Models-based Recommendation System (LRS) may introduce additional societal challenges to recommendation systems due to the inherent biases in Large Language Models (LLMs). From the perspective of item-side fairness, there remains a lack of comprehensive investigation into the item-side fairness of LRS given the unique characteristics of LRS compared to conventional recommendation systems. To bridge this gap, this study examines the property of LRS with respect to item-side fairness and reveals the influencing factors of both historical users' interactions and inherent semantic biases of LLMs, shedding light on the need to extend conventional item-side fairness methods for LRS. Towards this goal, we develop a concise and effective framework called IFairLRS to enhance the item-side fairness of an LRS. IFairLRS covers the main stages of building an LRS with specifically adapted strategies to calibrate the recommendations of LRS. We utilize IFairLRS to fine-tune LLaMA, a representative LLM, on \textit{MovieLens} and \textit{Steam} datasets, and observe significant item-side fairness improvements. The code can be found in https://github.com/JiangM-C/IFairLRS.git.

arXiv:2402.14903공식 코드 · Papers with Code

Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs

Tokenization, the division of input text into input tokens, is an often overlooked aspect of the large language model (LLM) pipeline and could be the source of useful or harmful inductive biases. Historically, LLMs have relied on byte pair encoding, without care to specific input domains. With the increased use of LLMs for reasoning, various number-specific tokenization schemes have been adopted, with popular models like LLaMa and PaLM opting for single-digit tokenization while GPT-3.5 and GPT-4 have separate tokens for each 1-, 2-, and 3-digit numbers. In this work, we study the effect this choice has on numerical reasoning through the use of arithmetic tasks. We consider left-to-right and right-to-left tokenization for GPT-3.5 and -4, finding that right-to-left tokenization (enforced by comma separating numbers at inference time) leads to largely improved performance. Furthermore, we find that model errors when using standard left-to-right tokenization follow stereotyped error patterns, suggesting that model computations are systematic rather than approximate. We show that the model is able to convert between tokenizations easily, thus allowing chain-of-thought-inspired approaches to recover performance on left-to-right tokenized inputs. We also find the gap between tokenization directions decreases when models are scaled, possibly indicating that larger models are better able to override this tokenization-dependent inductive bias. In summary, our work performs the first study of how number tokenization choices lead to differences in model performance on arithmetic tasks, accompanied by a thorough analysis of error patterns. We hope this work inspires practitioners to more carefully ablate number tokenization-related choices when working towards general models of numerical reasoning.