cs.CL（2025-04-26）

📊 共 9 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (6 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Theory of Mind in Large Language Models: Assessment and Enhancement	综述LLM心智理论能力：评估基准与提升策略分析	large language model
2	Detect, Explain, Escalate: Sustainable Dialogue Breakdown Management for LLM Agents	提出'检测、解释、升级'框架以解决对话中断问题	large language model chain-of-thought
3	Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks	LLM时代通用评估：超越基准的评测框架综述，关注能力评估与自动化	large language model instruction following
4	Graph of Attacks: Improved Black-Box and Interpretable Jailbreaks for LLMs	提出GoAT，利用攻击图提升LLM黑盒越狱攻击效果与可解释性	large language model	✅
5	LINC: Supporting Language Independent Communication and Comprehension to Enhance Contribution in Multilingual Collaborative Meetings	LINC：支持语言无关的沟通与理解，提升多语言协作会议的贡献度	multimodal
6	RAIR: Retrieval-Augmented Iterative Refinement for Chinese Spelling Correction	提出RAIR框架，增强LLM在中文拼写纠错中对领域术语和变长纠错的处理能力	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	🔗	⭐
7	KETCHUP: K-Step Return Estimation for Sequential Knowledge Distillation	提出KETCHUP：一种用于序列知识蒸馏的K步回报估计方法	reinforcement learning distillation large language model
8	SynLexLM: Scaling Legal LLMs with Synthetic Data and Curriculum Learning	SynLexLM：利用合成数据和课程学习扩展法律领域大语言模型	curriculum learning large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
9	Effective Length Extrapolation via Dimension-Wise Positional Embeddings Manipulation	提出维度感知位置编码操控(DPE)，无需训练即可有效外推LLM上下文长度。	manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页