cs.CL（2025-12-21）

📊 共 10 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) (7 🔗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) (7 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Emissions and Performance Trade-off Between Small and Large Language Models	研究表明，微调小型语言模型在多项任务上可实现与大型模型相当的性能，同时显著降低碳排放。	large language model
2	MDToC: Metacognitive Dynamic Tree of Concepts for Boosting Mathematical Problem-Solving of Large Language Models	提出MDToC，通过元认知动态概念树提升大语言模型数学问题求解能力	large language model
3	Rubric-Conditioned LLM Grading: Alignment, Uncertainty, and Robustness	提出基于规则条件的大语言模型评分框架，评估其对简答题的评分能力。	large language model
4	MemEvolve: Meta-Evolution of Agent Memory Systems	MemEvolve：通过元进化Agent记忆系统，提升Agent在复杂任务中的性能。	large language model
5	Solver-Independent Automated Problem Formulation via LLMs for High-Cost Simulation-Driven Design	APF：基于LLM的求解器无关自动化问题建模框架，用于高成本仿真驱动设计	large language model
6	A Comparative Study of Light-weight Language Models for PII Masking and their Deployment for Real Conversational Texts	轻量级语言模型用于PII掩码，在真实对话文本中性能可与大型模型媲美	large language model
7	LLMs on Drugs: Language Models Are Few-Shot Consumers	研究表明，LLM对推理时的人格设定敏感，提示词层面的“药物”干预会显著影响其性能。	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	🔗	⭐
8	From Word to World: Can Large Language Models be Implicit Text-based World Models?	探索LLM作为文本世界模型的潜力，提升Agent在文本环境中的学习效率	reinforcement learning world model large language model
9	LLM-CAS: Dynamic Neuron Perturbation for Real-Time Hallucination Correction	LLM-CAS：通过动态神经元扰动实现大语言模型实时幻觉纠正	reinforcement learning large language model multimodal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
10	Automated Red-Teaming Framework for Large Language Model Security Assessment: A Comprehensive Attack Generation and Detection System	提出自动化红队框架，用于大规模语言模型安全评估与漏洞挖掘	manipulation large language model chain-of-thought

⬅️ 返回 cs.CL 首页 · 🏠 返回主页