cs.CL（2024-10-10）

📊 共 52 篇论文 | 🔗 12 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (39 🔗9) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (39 篇)

#题目一句话要点标签🔗⭐
1 Sample then Identify: A General Framework for Risk Control and Assessment in Multimodal Large Language Models 提出TRON框架，用于多模态大语言模型风险控制与评估，提升开放环境下的可靠性。 large language model multimodal
2 Insight Over Sight: Exploring the Vision-Knowledge Conflicts in Multimodal LLMs 提出评估框架，揭示多模态大语言模型中视觉信息与常识知识的冲突问题。 large language model multimodal
3 What Makes Large Language Models Reason in (Multi-Turn) Code Generation? 研究提示工程对大语言模型多轮代码生成推理能力的影响 large language model chain-of-thought
4 Thought2Text: Text Generation from EEG Signal using Large Language Models (LLMs) Thought2Text：利用大语言模型从脑电信号生成文本，实现“意念转文字” large language model multimodal
5 Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models ECHOQA：探究大语言模型中参数知识与上下文知识的动态交互 large language model ✅
6 The Large Language Model GreekLegalRoBERTa 提出GreekLegalRoBERTa，提升希腊语法律文本的命名实体识别和主题分类性能。 large language model
7 VibeCheck: Discover and Quantify Qualitative Differences in Large Language Models VibeCheck：自动发现并量化大语言模型输出中的细微风格差异 large language model
8 A Closer Look at Machine Unlearning for Large Language Models 提出机器反学习方法以解决大语言模型中的隐私问题 large language model ✅
9 Can Knowledge Graphs Make Large Language Models More Trustworthy? An Empirical Study Over Open-ended Question Answering 提出OKGQA基准测试，评估知识图谱增强大语言模型在开放域问答中的可信度。 large language model ✅
10 Teaching-Inspired Integrated Prompting Framework: A Novel Approach for Enhancing Reasoning in Large Language Models 提出教学启发式集成提示框架，提升大语言模型算术推理能力 large language model ✅
11 Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models 提出Omni-MATH，一个面向大语言模型的奥林匹克级别数学推理基准 large language model
12 Disease Entity Recognition and Normalization is Improved with Large Language Model Derived Synthetic Normalized Mentions 利用LLM生成的合成归一化提及改进疾病实体识别与归一化 large language model
13 NusaMT-7B: Machine Translation for Low-Resource Indonesian Languages with Large Language Models NusaMT-7B：利用大型语言模型提升低资源印尼语机器翻译性能 large language model
14 Extracting and Combining Abilities For Building Multi-lingual Ability-enhanced Large Language Models 提出MAEC方法，无需训练即可为大语言模型赋予多语言能力。 large language model ✅
15 Uncovering Overfitting in Large Language Model Editing 揭示大语言模型知识编辑中的过拟合问题，并提出LTI方法缓解。 large language model
16 GameTraversalBenchmark: Evaluating Planning Abilities Of Large Language Models Through Traversing 2D Game Maps 提出GameTraversalBenchmark，评估大型语言模型在2D游戏地图中的规划能力。 large language model ✅
17 Detecting Training Data of Large Language Models via Expectation Maximization 提出EM-MIA方法以解决大语言模型训练数据检测问题 large language model
18 AI-Press: A Multi-Agent News Generating and Feedback Simulation System Powered by Large Language Models AI-Press：一个基于大语言模型的多智能体新闻生成与反馈模拟系统 large language model
19 OneNet: A Fine-Tuning Free Framework for Few-Shot Entity Linking via Large Language Model Prompting 提出OneNet，一种免微调的LLM提示框架，用于少样本实体链接。 large language model
20 Upcycling Large Language Models into Mixture of Experts 提出虚拟组初始化和权重缩放方法，高效地将大型语言模型升级为混合专家模型。 large language model
21 Diversity of Thought Elicits Stronger Reasoning Capabilities in Multi-Agent Debate Frameworks 多智能体辩论框架中，思维多样性增强LLM的推理能力 large language model chain-of-thought
22 Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning CLOVER：通过分解翻译和验证一阶逻辑，提升复杂逻辑推理能力。 large language model chain-of-thought
23 Advancing biomolecular understanding and design following human instructions InstructBioMol：通过自然语言指令驱动的生物分子理解与设计 large language model multimodal ✅
24 Dialectical Behavior Therapy Approach to LLM Prompting 提出基于辩证行为疗法的LLM提示方法，提升小模型推理能力 large language model chain-of-thought
25 MathCoder2: Better Math Reasoning from Continued Pretraining on Model-translated Mathematical Code MathCoder2：通过模型翻译的数学代码持续预训练，提升数学推理能力 large language model ✅
26 Responsible AI in NLP: GUS-Net Span-Level Bias Detection Dataset and Benchmark for Generalizations, Unfairness, and Stereotypes 提出GUS-Net框架，用于细粒度、可解释的NLP偏见检测与缓解。 large language model
27 From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions DRAFT框架：通过自驱动交互动态优化工具文档，提升LLM工具使用能力 large language model
28 COMPL-AI Framework: A Technical Interpretation and LLM Benchmarking Suite for the EU Artificial Intelligence Act COMPL-AI框架：欧盟AI法案的技术解读与LLM基准测试套件 large language model
29 Benchmarking Agentic Workflow Generation 提出WorfBench基准测试，评估LLM在复杂工作流生成中的能力，揭示序列与图规划能力差距。 large language model ✅
30 Semantic Self-Consistency: Enhancing Language Model Reasoning via Semantic Weighting 提出语义自洽性方法，通过语义加权增强语言模型推理能力 large language model
31 AgentBank: Towards Generalized LLM Agents via Fine-Tuning on 50000+ Interaction Trajectories AgentBank：通过在5万+交互轨迹上微调，实现通用LLM Agent large language model
32 Smart Audit System Empowered by LLM 提出基于LLM的智能审计系统，提升制造业质量审计效率与透明度。 large language model
33 No Free Lunch: Retrieval-Augmented Generation Undermines Fairness in LLMs, Even for Vigilant Users RAG增强大语言模型在提升性能的同时，可能损害公平性，即使对于有防范意识的用户。 large language model
34 RealVul: Can We Detect Vulnerabilities in Web Applications with LLM? RealVul：首个基于LLM的PHP漏洞检测框架，提升漏洞检测能力。 large language model
35 KRAG Framework for Enhancing LLMs in the Legal Domain 提出KRAG框架，通过知识表示增强LLM在法律领域的应用 large language model
36 MKGL: Mastery of a Three-Word Language 提出MKGL：使LLM掌握三词知识图谱语言，提升知识图谱补全精度。 large language model
37 DemoShapley: Valuation of Demonstrations for In-Context Learning 提出DemoShapley方法，通过Shapley值评估In-Context Learning中演示样本的贡献。 large language model
38 Using LLMs to Discover Legal Factors 利用大型语言模型自动发现法律领域关键因素，辅助法律分析 large language model
39 PublicHearingBR: A Brazilian Portuguese Dataset of Public Hearing Transcripts for Summarization of Long Documents 提出了PublicHearingBR，一个用于巴西葡萄牙语长文档摘要的公共听证会记录数据集。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
40 TPO: Aligning Large Language Models with Multi-branch & Multi-step Preference Trees TPO：通过多分支多步偏好树对齐大型语言模型，提升复杂推理能力 preference learning DPO direct preference optimization ✅
41 StablePrompt: Automatic Prompt Tuning using Reinforcement Learning for Large Language Models StablePrompt：利用强化学习自动调整大语言模型提示，提升训练稳定性和性能。 reinforcement learning large language model
42 Privacy-preserved LLM Cascade via CoT-enhanced Policy Learning 提出P³Defer以解决隐私保护的LLM级联问题 policy learning large language model chain-of-thought
43 Rewriting Conversational Utterances with Instructed Large Language Models 利用指令式大型语言模型重写对话语句，提升会话搜索效果。 reinforcement learning large language model
44 StepTool: Enhancing Multi-Step Tool Usage in LLMs via Step-Grained Reinforcement Learning StepTool：通过步粒度强化学习增强LLM的多步工具使用能力 reinforcement learning reward shaping large language model ✅
45 Increasing the Difficulty of Automatically Generated Questions via Reinforcement Learning with Synthetic Preference 提出基于强化学习与合成偏好的问题生成方法，提升文化遗产领域机器阅读理解数据集难度。 reinforcement learning PPO RLHF
46 How Does Vision-Language Adaptation Impact the Safety of Vision Language Models? 研究视觉语言适配对视觉语言模型安全性的影响，并提出权重融合方法。 reinforcement learning large language model multimodal
47 Exploring Natural Language-Based Strategies for Efficient Number Learning in Children through Reinforcement Learning 利用强化学习探索基于自然语言的儿童高效数字学习策略 reinforcement learning deep reinforcement learning
48 Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System Optima：优化LLM多智能体系统的效率与效能 DPO direct preference optimization large language model ✅
49 GenARM: Reward Guided Generation with Autoregressive Reward Model for Test-time Alignment GenARM：利用自回归奖励模型进行测试时对齐的奖励引导生成 reinforcement learning large language model
50 RecurFormer: Not All Transformer Heads Need Self-Attention RecurFormer：用线性循环网络替换Transformer中冗余自注意力头，提升长文本推理效率。 Mamba large language model
51 Multi-Facet Counterfactual Learning for Content Quality Evaluation 提出MOLE框架，利用多方面反事实学习提升内容质量评估的准确性 contrastive learning large language model
52 MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimization 提出MACPO框架，通过多智能体对比偏好优化实现弱监督到强模型的对齐。 RLHF large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页