cs.CL（2025-10-26）

📊 共 21 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Leveraging Large Language Models to Identify Conversation Threads in Collaborative Learning	利用大型语言模型识别协作学习中的对话主题，提升会话分析性能。	large language model
2	MMPersuade: A Dataset and Evaluation Framework for Multimodal Persuasion	MMPersuade：多模态说服数据集与评估框架，用于评估大型视觉语言模型的说服力	multimodal
3	Low-Resource Dialect Adaptation of Large Language Models: A French Dialect Case-Study	提出低资源方言自适应方法，利用参数高效微调提升LLM在魁北克法语上的性能	large language model
4	Frustratingly Easy Task-aware Pruning for Large Language Models	提出任务感知剪枝方法，在压缩大语言模型的同时保持特定任务性能。	large language model
5	Once Upon an Input: Reasoning via Per-Instance Program Synthesis	提出Per-Instance Program Synthesis (PIPS)方法，提升LLM在复杂推理任务中的性能。	large language model chain-of-thought
6	A Comprehensive Dataset for Human vs. AI Generated Text Detection	构建大规模人机生成文本检测数据集，促进AI生成内容溯源与鉴别	large language model	✅
7	Interpreting and Mitigating Unwanted Uncertainty in LLMs	探究并缓解大型语言模型中不期望的答案不确定性现象	large language model
8	Cross-Lingual Stability and Bias in Instruction-Tuned Language Models for Humanitarian NLP	针对人道主义NLP，评估指令调优语言模型在跨语言稳定性与偏差上的表现。	large language model
9	EchoMind: An Interrelated Multi-level Benchmark for Evaluating Empathetic Speech Language Models	提出EchoMind：一个多层次关联的基准，用于评估具身同理心的语音语言模型	instruction following
10	Beyond Semantics: How Temporal Biases Shape Retrieval in Transformer and State-Space Models	研究Transformer和状态空间模型中的时间偏差对上下文学习检索的影响	large language model
11	SALSA: Single-pass Autoregressive LLM Structured Classification	SALSA：单次自回归LLM结构化分类方法，提升文本分类性能	large language model
12	Rule-Based Explanations for Retrieval-Augmented LLM Systems	提出基于规则的解释方法，用于增强检索的大语言模型系统，提升可解释性。	large language model
13	AutoBench: Automating LLM Evaluation through Reciprocal Peer Assessment	AutoBench：通过互惠互评自动评估大型语言模型	large language model
14	Pedagogy-driven Evaluation of Generative AI-powered Intelligent Tutoring Systems	针对生成式AI驱动的智能辅导系统，提出教学法驱动的评估框架	large language model
15	SABlock: Semantic-Aware KV Cache Eviction with Adaptive Compression Block Size	SABlock：基于语义感知的自适应块大小KV缓存淘汰，提升长文本LLM推理效率	large language model
16	LooGLE v2: Are LLMs Ready for Real World Long Dependency Challenges?	LooGLE v2：评估LLM在真实世界长依赖任务中的能力，揭示其局限性。	large language model
17	Text to Trust: Evaluating Fine-Tuning and LoRA Trade-offs in Language Models for Unfair Terms of Service Detection	评估语言模型在不公平服务条款检测中微调与LoRA的权衡	large language model
18	CHOIR: Collaborative Harmonization fOr Inference Robustness	提出CHOIR，通过协同多个角色化LLM推理信号，提升推理鲁棒性。	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	🔗	⭐
19	Scalable Supervising Software Agents with Patch Reasoner	提出R4P补丁验证模型，通过推理实现软件工程Agent训练的可扩展监督。	reinforcement learning large language model
20	Adaptive Testing for LLM Evaluation: A Psychometric Alternative to Static Benchmarks	提出ATLAS自适应测试框架，高效评估大语言模型能力。	MAE large language model	✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
21	Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation	Ming-UniAudio：统一语音表示的语音LLM，用于联合理解、生成和编辑	manipulation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页