cs.CL（2024-10-21）

📊 共 47 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (41 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (41 篇)

#题目一句话要点标签🔗⭐
1 Pangea: A Fully Open Multilingual Multimodal LLM for 39 Languages Pangea：一个面向39种语言的完全开放的多语言多模态大语言模型 large language model multimodal
2 A Theoretical Understanding of Chain-of-Thought: Coherent Reasoning and Error-Aware Demonstration 理论分析链式思考：连贯推理与误差感知演示提升LLM性能 large language model chain-of-thought
3 AMPLE: Emotion-Aware Multimodal Fusion Prompt Learning for Fake News Detection 提出AMPLE框架，融合情感信息与多模态提示学习，提升假新闻检测性能 large language model multimodal ✅
4 Comparative Study of Multilingual Idioms and Similes in Large Language Models 对比研究大型语言模型在多语言隐喻和明喻理解中的表现 large language model chain-of-thought
5 Resource-Efficient Medical Report Generation using Large Language Models 提出一种资源高效的医学报告生成框架，利用视觉大语言模型提升报告质量。 large language model
6 1024m at SMM4H 2024: Tasks 3, 5 & 6 -- Ensembles of Transformers and Large Language Models for Medical Text Classification 利用Transformer和LLM集成模型解决SMM4H 2024医疗文本分类任务 large language model
7 Large Language Models for Cross-lingual Emotion Detection 利用大型语言模型及集成方法进行跨语言情感检测 large language model
8 Self-Explained Keywords Empower Large Language Models for Code Generation 提出自解释关键词（SEK）方法，提升大语言模型在代码生成中对低频关键词的理解。 large language model
9 Do Large Language Models Have an English Accent? Evaluating and Improving the Naturalness of Multilingual LLMs 提出多语言LLM自然度评测指标与对齐方法，提升非英语生成质量。 large language model
10 Who's Who: Large Language Models Meet Knowledge Conflicts in Practice 提出WhoQA基准数据集，用于评估大语言模型在知识冲突场景下的表现 large language model
11 DocEdit-v2: Document Structure Editing Via Multimodal LLM Grounding DocEdit-v2：提出一种基于多模态LLM的文档结构编辑框架，提升文档编辑性能。 multimodal
12 ToW: Thoughts of Words Improve Reasoning in Large Language Models 提出词语思考(ToW)数据增强方法，提升大语言模型推理能力并减少幻觉。 large language model
13 Large Language Models Know What To Say But Not When To Speak 提出包含内转折过渡相关位置标注的数据集，评估大语言模型在口语对话中预测时机的能力。 large language model
14 Exploring Continual Fine-Tuning for Enhancing Language Ability in Large Language Model 提出持续微调方法以提升大语言模型的语言能力 large language model
15 Did somebody say "Gest-IT"? A pilot exploration of multimodal data management Gest-IT：构建多模态语料库，探索视力正常人与视障人士对话中的手势模式差异。 multimodal
16 Mitigating Hallucinations of Large Language Models in Medical Information Extraction via Contrastive Decoding 提出ALCD以解决医疗信息提取中的幻觉问题 large language model
17 GATEAU: Selecting Influential Samples for Long Context Alignment GATEAU：通过选择关键样本提升长文本对齐能力 large language model instruction following
18 Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Following 提出Multi-IF基准，评估LLM在多轮和多语言指令跟随方面的能力 large language model instruction following
19 Improving Neuron-level Interpretability with White-box Language Models 提出CRATE：一种白盒Transformer架构，提升神经元级可解释性 foundation model
20 MagicPIG: LSH Sampling for Efficient LLM Generation MagicPIG：基于LSH采样的LLM高效生成方法，提升长文本处理性能。 large language model ✅
21 RAC: Efficient LLM Factuality Correction with Retrieval Augmentation 提出检索增强校正(RAC)方法，高效提升大语言模型的事实性准确度。 large language model ✅
22 Scalable Data Ablation Approximations for Language Models through Modular Training and Merging 提出基于模块化训练和模型合并的可扩展数据消融近似方法，加速LLM数据评估。 large language model
23 Stacking Small Language Models for Generalizability 提出FSLM：堆叠小型语言模型以提升通用性，降低训练与推理成本 large language model
24 Catastrophic Failure of LLM Unlearning via Quantization 量化揭示LLM卸载学习的灾难性失败：模型遗忘实为隐藏 large language model ✅
25 CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolution 提出CompassJudger-1：首个开源一体化评判LLM，用于模型评估与演进。 large language model ✅
26 RULEBREAKERS: Challenging LLMs at the Crossroads between Formal Logic and Human-like Reasoning 提出RULEBREAKERS数据集，揭示LLM在形式逻辑与类人推理的交叉点上的局限性 large language model
27 To the Globe (TTG): Towards Language-Driven Guaranteed Travel Planning TTG：提出一种语言驱动的保证性旅行规划系统，解决复杂旅行安排问题。 large language model
28 Can Knowledge Editing Really Correct Hallucinations? 提出HalluEditBench，用于评估知识编辑方法在纠正大语言模型幻觉方面的能力。 large language model
29 Building A Coding Assistant via the Retrieval-Augmented Language Model 提出CONAN：一种检索增强的语言模型，用于构建代码助手 large language model
30 Contamination Report for Multilingual Benchmarks 研究揭示大型语言模型在多语言基准测试中普遍存在的污染问题 large language model
31 Exploring Pretraining via Active Forgetting for Improving Cross Lingual Transfer for Decoder Language Models 提出基于主动遗忘的预训练方法，提升解码器语言模型跨语言迁移能力。 large language model
32 A Troublemaker with Contagious Jailbreak Makes Chaos in Honest Towns 提出TMCHT框架与ARCJ方法，评估并提升多智能体系统中对抗性攻击的有效性 large language model
33 1-bit AI Infra: Part 1.1, Fast and Lossless BitNet b1.58 Inference on CPUs bitnet.cpp：加速CPU上无损BitNet b1.58推理的定制化软件栈 large language model ✅
34 A Psycholinguistic Evaluation of Language Models' Sensitivity to Argument Roles 评估语言模型对论元角色敏感性的心理语言学研究 large language model
35 Can Large Audio-Language Models Truly Hear? Tackling Hallucinations with Multi-Task Assessment and Stepwise Audio Reasoning 提出多任务评估与逐步音频推理，解决大型音频语言模型中的幻觉问题 chain-of-thought
36 Do LLMs write like humans? Variation in grammatical and rhetorical styles 通过语法和修辞风格的差异，揭示大型语言模型与人类写作的本质区别 large language model
37 Analysing the Residual Stream of Language Models Under Knowledge Conflicts 通过分析LLM残差流，检测知识冲突并预测模型行为 large language model
38 Surprise! Uniform Information Density Isn't the Whole Story: Predicting Surprisal Contours in Long-form Discourse 提出结构化上下文假设，预测长文本语篇中的Surprisal轮廓，超越均匀信息密度理论。 large language model
39 CausalGraph2LLM: Evaluating LLMs for Causal Queries CausalGraph2LLM：评估大型语言模型在因果查询中的能力 large language model
40 Guardians of Discourse: Evaluating LLMs on Multilingual Offensive Language Detection 评估LLM在多语言攻击性语言检测中的表现，揭示其偏见与局限性 large language model
41 A Survey of Conversational Search 综述性论文：全面解析会话式搜索技术，展望未来发展方向 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
42 Can Large Language Models Invent Algorithms to Improve Themselves?: Algorithm Discovery for Recursive Self-Improvement through Reinforcement Learning Self-Developing：LLM自主发现算法实现递归自提升，超越人工设计 reinforcement learning direct preference optimization large language model
43 Pre-training Distillation for Large Language Models: A Design Space Exploration 探索预训练蒸馏设计空间，提升大语言模型知识迁移效率 distillation large language model
44 Enhancing Multimodal Affective Analysis with Learned Live Comment Features 提出LCAffect数据集并利用对比学习生成合成弹幕特征，提升多模态情感分析性能 contrastive learning multimodal
45 Revealing and Mitigating the Local Pattern Shortcuts of Mamba 揭示并缓解Mamba模型局部模式捷径问题，提升长程依赖处理能力 Mamba SSM state space model
46 R2Gen-Mamba: A Selective State Space Model for Radiology Report Generation 提出R2Gen-Mamba，利用选择性状态空间模型高效生成放射科报告。 Mamba state space model
47 RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style 提出RM-Bench，评估奖励模型对语言模型细微内容差异和风格偏差的敏感性。 reinforcement learning RLHF ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页