cs.CL（2025-03-07）

📊 共 25 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 Can Large Language Models Simulate Human Responses? A Case Study of Stated Preference Experiments in the Context of Heating-related Choices 利用大型语言模型模拟人类在供暖选择偏好实验中的行为 large language model chain-of-thought
2 S2S-Arena, Evaluating Speech2Speech Protocols on Instruction Following with Paralinguistic Information 提出S2S-Arena：一个评估语音到语音模型指令跟随能力及副语言信息的基准 large language model instruction following
3 Dynamic Knowledge Integration for Evidence-Driven Counter-Argument Generation with Large Language Models 提出动态知识集成方法，提升大语言模型生成证据驱动的反驳论证质量 large language model
4 Evaluating open-source Large Language Models for automated fact-checking 评估开源大语言模型在自动化事实核查中的能力与局限性 large language model
5 Chain of Strategy Optimization Makes Large Language Models Better Emotional Supporter 提出策略链优化方法，提升大语言模型的情感支持能力 large language model
6 Coreference as an indicator of context scope in multimodal narrative 揭示多模态叙事中大型语言模型与人类指代消解模式的差异 multimodal ✅
7 AdaSpec: Adaptive Speculative Decoding for Fast, SLO-Aware Large Language Model Serving AdaSpec：自适应推测解码，加速LLM服务并满足SLO large language model ✅
8 Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching 提出Sketch-of-Thought，通过认知启发式草图方法提升LLM推理效率并减少token使用。 large language model chain-of-thought
9 RocketEval: Efficient Automated LLM Evaluation via Grading Checklist RocketEval：通过检查清单分级实现高效的自动化LLM评估 large language model chain-of-thought ✅
10 IDEA Prune: An Integrated Enlarge-and-Prune Pipeline in Generative Language Model Pretraining 提出IDEA Prune，一种生成语言模型预训练中集成的放大-剪枝流水线，提升剪枝模型性能。 large language model
11 QG-SMS: Enhancing Test Item Analysis via Student Modeling and Simulation 提出QG-SMS框架，利用学生建模与模拟增强试题分析，提升问题生成评估质量 large language model
12 WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs 提出WikiBigEdit大规模基准，评估LLM终身知识编辑的极限 large language model
13 ORANSight-2.0: Foundational LLMs for O-RAN ORANSight-2.0：为O-RAN定制的基础大语言模型，提升特定任务性能。 large language model
14 SINdex: Semantic INconsistency Index for Hallucination Detection in LLMs 提出SINdex，一种基于语义不一致性的LLM幻觉检测方法 large language model
15 SANDWiCH: Semantical Analysis of Neighbours for Disambiguating Words in Context ad Hoc SANDWiCH：提出一种基于邻域语义分析的多语言词义消歧框架，达到新的SOTA。 large language model
16 This Is Your Doge, If It Please You: Exploring Deception and Robustness in Mixture of LLMs 揭示LLM混合架构的欺骗脆弱性，并提出防御机制以提升鲁棒性 large language model
17 AutoIOT: LLM-Driven Automated Natural Language Programming for AIoT Applications AutoIOT：LLM驱动的AIoT应用自然语言自动编程 large language model
18 Knowledge Updating? No More Model Editing! Just Selective Contextual Reasoning 提出选择性上下文推理(SCR)，无需模型编辑即可更新LLM知识。 large language model
19 Rewarding Curse: Analyze and Mitigate Reward Modeling Issues for LLM Reasoning 针对LLM推理中奖励模型问题的分析与缓解方法 chain-of-thought
20 EMCee: Improving Multilingual Capability of LLMs via Bridging Knowledge and Reasoning with Extracted Synthetic Multilingual Context EMCee：通过提取合成多语言上下文桥接知识与推理，提升LLM的多语言能力 large language model
21 DETQUS: Decomposition-Enhanced Transformers for QUery-focused Summarization DETQUS：利用分解增强Transformer，解决查询聚焦的表格摘要生成问题 large language model
22 MastermindEval: A Simple But Scalable Reasoning Benchmark 提出MastermindEval：一个简单可扩展的推理基准测试，用于评估大型语言模型的演绎推理能力。 large language model
23 MM-StoryAgent: Immersive Narrated Storybook Video Generation with a Multi-Agent Paradigm across Text, Image and Audio MM-StoryAgent：提出一种多智能体框架，用于生成沉浸式叙事故事书视频 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
24 Learning LLM Preference over Intra-Dialogue Pairs: A Framework for Utterance-level Understandings 提出基于LLM偏好学习的对话单元理解框架，提升小模型性能 preference learning large language model
25 Similarity-Based Domain Adaptation with LLMs 提出基于LLM相似度的领域自适应框架，无需源域模型训练即可提升跨域文本分类性能。 distillation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页