cs.CL（2024-06-27）

📊 共 40 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (35 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (35 篇)

#题目一句话要点标签🔗⭐
1 DIM: Dynamic Integration of Multimodal Entity Linking with Large Language Model 提出DIM方法，利用大语言模型动态融合多模态信息，提升实体链接性能。 large language model multimodal ✅
2 Rethinking harmless refusals when fine-tuning foundation models 提出基于理由的欺骗现象，并验证反驳比拒绝更能有效抑制有害行为 large language model foundation model chain-of-thought
3 FlowVQA: Mapping Multimodal Logic in Visual Question Answering with Flowcharts FlowVQA：提出一个基于流程图的多模态视觉问答新基准，用于评估模型的逻辑推理能力。 multimodal visual grounding
4 Fairness and Bias in Multimodal AI: A Survey 调查多模态AI中的公平性和偏见问题，强调预处理缓解方法的重要性。 large language model multimodal
5 Revealing Fine-Grained Values and Opinions in Large Language Models 通过分析LLM对政治倾向测试的响应，揭示其潜在价值观和偏见。 large language model
6 Adaptive Draft-Verification for Efficient Large Language Model Decoding 提出自适应草稿验证ADED，加速大语言模型解码且无需微调。 large language model
7 Data Generation Using Large Language Models for Text Classification: An Empirical Case Study 利用大型语言模型生成合成数据用于文本分类的实证研究 large language model
8 Captioning Visualizations with Large Language Models (CVLLM): A Tutorial 利用大型语言模型自动生成可视化图表的标题，探索InfoVis领域的新可能。 large language model
9 Inclusivity in Large Language Models: Personality Traits and Gender Bias in Scientific Abstracts 评估大型语言模型在科学摘要中的包容性：人格特质与性别偏见分析 large language model
10 Statements: Universal Information Extraction from Tables with Large Language Models for ESG KPIs 提出Statements结构和SemTabNet数据集，利用大语言模型从表格中提取ESG KPI信息。 large language model
11 Follow-Up Questions Improve Documents Generated by Large Language Models 通过后续问题提升大型语言模型生成文档的质量 large language model
12 Can Large Language Models Generate High-quality Patent Claims? 探索大语言模型在专利权利要求生成中的能力，并分析其优劣势 large language model
13 The Model Arena for Cross-lingual Sentiment Analysis: A Comparative Study in the Era of Large Language Models 跨语言情感分析模型竞技场：大型语言模型时代的对比研究 large language model
14 STBench: Assessing the Ability of Large Language Models in Spatio-Temporal Analysis STBench：评估大语言模型在时空分析中的能力 large language model ✅
15 DataGen: Unified Synthetic Dataset Generation via Large Language Models DataGen：提出一种基于大语言模型的统一合成数据集生成框架，提升数据质量与可控性。 large language model
16 SSP: Self-Supervised Prompting for Cross-Lingual Transfer to Low-Resource Languages using Large Language Models 提出自监督提示SSP，利用大语言模型实现低资源语言的跨语言迁移 large language model
17 OutlierTune: Efficient Channel-Wise Quantization for Large Language Models OutlierTune：面向大语言模型的高效通道量化方法，提升INT6量化精度与效率。 large language model
18 Two-Pronged Human Evaluation of ChatGPT Self-Correction in Radiology Report Simplification 利用自校正提示，ChatGPT在放射报告简化任务中表现出色 large language model chain-of-thought
19 DeSTA: Enhancing Speech Language Models through Descriptive Speech-Text Alignment 提出描述性语音-文本对齐方法DeSTA，增强语音语言模型对语音非语言特征的理解和泛化能力。 large language model instruction following
20 DiVERT: Distractor Generation with Variational Errors Represented as Text for Math Multiple-choice Questions DiVERT：基于变分误差文本表示的数学多选题干扰项生成 large language model
21 T-FREE: Subword Tokenizer-Free Generative LLMs via Sparse Representations for Memory-Efficient Embeddings T-FREE：通过稀疏表示实现内存高效嵌入的无子词分词器生成式LLM large language model
22 LongLaMP: A Benchmark for Personalized Long-form Text Generation 提出LongLaMP基准，用于评估个性化长文本生成任务 large language model
23 Leveraging Machine-Generated Rationales to Facilitate Social Meaning Detection in Conversations 利用机器生成的推理增强对话理解，提升社交含义检测性能 large language model
24 Hierarchical Deconstruction of LLM Reasoning: A Graph-Based Framework for Analyzing Knowledge Utilization 提出基于图的LLM推理分层解构框架，分析知识利用方式 large language model
25 Development and Evaluation of a Retrieval-Augmented Generation Tool for Creating SAPPhIRE Models of Artificial Systems 提出基于检索增强生成（RAG）的工具，用于创建人工系统SAPPhIRE模型 large language model
26 AutoRAG-HP: Automatic Online Hyper-Parameter Tuning for Retrieval-Augmented Generation 提出AutoRAG-HP以解决RAG系统的超参数优化问题 large language model
27 EmPO: Emotion Grounding for Empathetic Response Generation through Preference Optimization EmPO：通过偏好优化和情感基础提升共情回复生成 large language model ✅
28 LLM-based Frameworks for API Argument Filling in Task-Oriented Conversational Systems 提出基于LLM的API参数填充框架，提升任务型对话系统性能 large language model
29 Building Understandable Messaging for Policy and Evidence Review (BUMPER) with AI BUMPER框架利用AI构建可理解的消息传递系统，用于政策和证据审查 large language model
30 Does ChatGPT Have a Mind? 探讨大型语言模型是否具备心智，聚焦其是否拥有信念、欲望和意图 large language model
31 Are Generative Language Models Multicultural? A Study on Hausa Culture and Emotions using ChatGPT 评估ChatGPT在豪萨文化和情感理解上的表现，揭示其在低资源语言文化适应性的局限性 large language model
32 Improving Weak-to-Strong Generalization with Reliability-Aware Alignment 提出可靠性感知对齐方法，提升弱监督到强模型的泛化能力 large language model ✅
33 Capturing Minds, Not Just Words: Enhancing Role-Playing Language Models with Personality-Indicative Data 通过个性化数据增强角色扮演语言模型的表现 large language model ✅
34 Sonnet or Not, Bot? Poetry Evaluation for Large Models and Datasets 评估大型语言模型诗歌形式识别能力，揭示模型对诗歌特征的理解程度 large language model
35 Can we teach language models to gloss endangered languages? 利用大型语言模型和上下文学习，实现濒危语言的自动词间对齐标注。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
36 Suri: Multi-constraint Instruction Following for Long-form Text Generation Suri：面向长文本生成的多约束指令跟随数据集与I-ORPO对齐方法 DPO instruction following ✅
37 Aligning Teacher with Student Preferences for Tailored Training Data Generation ARTE：对齐教师与学生偏好，定制化生成知识蒸馏训练数据 distillation large language model
38 Simulating Classroom Education with LLM-Empowered Agents SimClass：基于LLM驱动的多智能体模拟课堂教学框架 teacher-student large language model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
39 Read Anywhere Pointed: Layout-aware GUI Screen Reading with Tree-of-Lens Grounding 提出基于Tree-of-Lens的GUI屏幕阅读代理，解决用户指定位置的屏幕内容理解问题。 spatial relationship large language model multimodal

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
40 LiveBench: A Challenging, Contamination-Limited LLM Benchmark LiveBench：一个具有挑战性且污染受限的大语言模型评测基准 AMP instruction following

⬅️ 返回 cs.CL 首页 · 🏠 返回主页