cs.CL（2024-06-19）

📊 共 47 篇论文 | 🔗 10 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (42 🔗9) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (42 篇)

#题目一句话要点标签🔗⭐
1 Transferable speech-to-text large language model alignment module 提出可迁移的语音到文本大语言模型对齐模块，简化多模态任务架构。 large language model foundation model
2 Towards Robust Evaluation: A Comprehensive Taxonomy of Datasets and Metrics for Open Domain Question Answering in the Era of Large Language Models 构建开放域问答数据集与评估指标的综合分类体系，促进大语言模型时代下的鲁棒评估。 large language model multimodal
3 Knowledge Graph-Enhanced Large Language Models via Path Selection 提出KELP框架，通过路径选择增强知识图谱赋能的大语言模型，提升事实准确性。 large language model
4 Optimizing Psychological Counseling with Instruction-Tuned Large Language Models 利用指令调优的大语言模型优化心理咨询 large language model
5 Open Generative Large Language Models for Galician 提出面向加利西亚语的开源生成式大语言模型，提升小语种NLP技术可及性。 large language model
6 Adaptable Logical Control for Large Language Models Ctrl-G：一种可控的大语言模型生成框架，通过HMM实现逻辑约束 large language model
7 Evaluating Large Language Models along Dimensions of Language Variation: A Systematik Invesdigatiom uv Cross-lingual Generalization 通过可控的语言变异建模，系统性评估大语言模型的跨语言泛化能力。 large language model
8 Leveraging Large Language Models to Measure Gender Representation Bias in Gendered Language Corpora 提出一种基于LLM的方法，用于检测和量化性别化语言语料库中的性别表征偏差。 large language model
9 Jailbreaking Large Language Models Through Alignment Vulnerabilities in Out-of-Distribution Settings 提出ObscurePrompt方法，利用分布外数据脆弱性破解大语言模型对齐限制 large language model
10 In-Context Former: Lightning-fast Compressing Context for Large Language Model 提出IC-Former，通过线性复杂度上下文压缩加速大语言模型推理。 large language model
11 ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models 提出ZeroDL，利用大语言模型实现文本聚类的零样本分布学习 large language model
12 BeHonest: Benchmarking Honesty in Large Language Models 提出BeHonest基准以评估大型语言模型的诚实性问题 large language model ✅
13 Locating and Extracting Relational Concepts in Large Language Models 提出基于因果中介分析的关系概念定位方法，并成功从LLM中提取关系表示。 large language model
14 Large Language Models are Biased Because They Are Large Language Models 大型语言模型固有的设计导致其不可避免地产生偏差 large language model
15 PathoLM: Identifying pathogenicity from the DNA sequence through the Genome Foundation Model PathoLM：利用基因组基础模型从DNA序列中识别病原体 foundation model
16 Improving Visual Commonsense in Language Models via Multiple Image Generation 提出多图生成方法以提升语言模型的视觉常识推理能力 large language model multimodal ✅
17 On the Utility of Domain-Adjacent Fine-Tuned Model Ensembles for Few-shot Problems 提出DAFT-E框架，利用领域邻近微调模型集成解决少样本问题 large language model foundation model
18 Finding Blind Spots in Evaluator LLMs with Interpretable Checklists 提出FBI框架，揭示评估LLM在事实性、推理等能力评估上的盲点。 large language model instruction following ✅
19 LIVE: Learnable In-Context Vector for Visual Question Answering 提出LIVE：一种可学习的上下文向量，用于提升视觉问答任务中的上下文学习能力。 large language model multimodal ✅
20 VDebugger: Harnessing Execution Feedback for Debugging Visual Programs VDebugger：利用执行反馈调试视觉程序，提升视觉推理准确性 large language model ✅
21 WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia WikiContradict：一个评估LLM在维基百科知识冲突处理能力的基准 large language model
22 Learn and Unlearn: Addressing Misinformation in Multilingual LLMs 提出多语言LLM的有害信息传播与消除方法，解决跨语言污染问题 large language model
23 Multi-View Empowered Structural Graph Wordification for Language Models 提出Dr.E框架，实现图结构数据与大语言模型的token级对齐。 large language model ✅
24 Developing Story: Case Studies of Generative AI's Use in Journalism 揭示新闻机构使用生成式AI的案例研究，强调记者与LLM互动中的敏感信息处理与内容生成风险。 large language model
25 Distributional reasoning in LLMs: Parallel reasoning processes in multi-hop reasoning 提出一种可解释的LLM多跳推理分析方法，揭示模型内部的并行推理过程 large language model
26 LLMs as Models for Analogical Reasoning 利用大型语言模型进行类比推理建模，探索其认知能力 large language model
27 Can LLMs Reason in the Wild with Programs? 提出“野外推理”任务，揭示LLM在复杂开放场景下的推理局限性 large language model
28 DoubleDipper: Improving Long-Context LLMs via Context Recycling DoubleDipper：通过上下文回收提升长文本LLM的问答性能 large language model
29 Dual-Phase Accelerated Prompt Optimization 提出双阶段加速Prompt优化方法，提升闭源大语言模型在多任务上的性能。 large language model
30 SQLFixAgent: Towards Semantic-Accurate Text-to-SQL Parsing via Consistency-Enhanced Multi-Agent Collaboration 提出SQLFixAgent，通过一致性增强的多Agent协作提升Text-to-SQL语义准确性 large language model
31 ALiiCE: Evaluating Positional Fine-grained Citation Generation 提出ALiiCE框架，用于评估LLM在句子内位置粒度上的引文生成质量 large language model
32 SD-Eval: A Benchmark Dataset for Spoken Dialogue Understanding Beyond Words SD-Eval：一个用于评估语音对话理解中超词汇信息的基准数据集 large language model ✅
33 Improving Zero-shot LLM Re-Ranker with Risk Minimization 提出UR^3框架以降低零-shot LLM重排序中的估计偏差 large language model
34 R^2AG: Incorporating Retrieval Information into Retrieval Augmented Generation 提出R^2AG以解决LLMs与检索器之间的语义差距问题 large language model
35 Data Contamination Can Cross Language Barriers 揭示并防御LLM中跨语言数据污染，提升模型泛化能力 large language model ✅
36 Probing the Emergence of Cross-lingual Alignment during LLM Training 利用神经元探针揭示LLM训练中跨语言对齐的涌现机制 large language model
37 Automating IRAC Analysis in Malaysian Contract Law using a Semi-Structured Knowledge Base 提出LegalSemi基准和结构化知识库，提升LLM在马来西亚合同法IRAC分析中的表现 large language model
38 Multi-Meta-RAG: Improving RAG for Multi-Hop Queries using Database Filtering with LLM-Extracted Metadata 提出Multi-Meta-RAG，利用LLM提取元数据进行数据库过滤，提升多跳查询RAG性能 large language model ✅
39 Synthetic Context Generation for Question Generation 提出基于LLM合成上下文的问题生成方法，提升小模型性能 large language model
40 DialSim: A Dialogue Simulator for Evaluating Long-Term Multi-Party Dialogue Understanding of Conversational Agents DialSim：用于评估会话代理长期多方对话理解的对话模拟器 large language model
41 When Parts Are Greater Than Sums: Individual LLM Components Can Outperform Full Models 通过重加权LLM内部组件，提升小样本学习分类任务性能 large language model
42 Learning to Generate Answers with Citations via Factual Consistency Models 提出基于事实一致性模型的弱监督微调方法，提升LLM生成答案时引用准确性。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
43 Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models 提出AutoIF，通过执行反馈自博弈提升大语言模型的指令跟随能力 reinforcement learning RLHF DPO ✅
44 BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation 提出BiLD损失，通过双向Logits差异蒸馏提升大语言模型性能。 distillation large language model
45 Enhancing Automated Audio Captioning via Large Language Models with Optimized Audio Encoding 利用优化音频编码和大型语言模型增强自动音频字幕生成 distillation large language model
46 Multi-Stage Balanced Distillation: Addressing Long-Tail Challenges in Sequence-Level Knowledge Distillation 提出多阶段平衡蒸馏框架，解决序列级知识蒸馏中的长尾分布问题 distillation large language model

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

#题目一句话要点标签🔗⭐
47 GSR-BENCH: A Benchmark for Grounded Spatial Reasoning Evaluation via Multimodal LLMs GSR-BENCH：通过多模态LLM评估具身空间推理的基准 spatial relationship multimodal

⬅️ 返回 cs.CL 首页 · 🏠 返回主页