cs.CL（2025-12-09）

📊 共 11 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) 支柱三：空间感知与语义 (Perception & Semantics) (1) 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 Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detection Using Large Language Models 提出基于显式推理视角的软归纳偏置方法，用于不当言论检测 large language model chain-of-thought
2 Knowledge-Guided Large Language Model for Automatic Pediatric Dental Record Understanding and Safe Antibiotic Recommendation 提出知识引导的大型语言模型以解决儿童牙科记录理解和安全抗生素推荐问题 large language model
3 QSTN: A Modular Framework for Robust Questionnaire Inference with Large Language Models QSTN：一个用于大型语言模型稳健问卷推断的模块化框架 large language model
4 MindShift: Analyzing Language Models' Reactions to Psychological Prompts MindShift：分析语言模型对心理学提示的反应，评估其人格适应性 large language model
5 Detecting Hallucinations in Graph Retrieval-Augmented Generation via Attention Patterns and Semantic Alignment 提出GGA模型，通过注意力模式和语义对齐检测GraphRAG中的幻觉问题 large language model
6 Toward Faithful Retrieval-Augmented Generation with Sparse Autoencoders 提出RAGLens，利用稀疏自编码器提升检索增强生成的事实性检测。 large language model ✅
7 HarmTransform: Transforming Explicit Harmful Queries into Stealthy via Multi-Agent Debate 提出HarmTransform，通过多智能体辩论将显式有害查询转化为隐蔽性更强的形式。 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 A Systematic Evaluation of Preference Aggregation in Federated RLHF for Pluralistic Alignment of LLMs 提出联邦RLHF中自适应偏好聚合方法，提升LLM对多元化人类偏好对齐的公平性 PPO RLHF large language model
9 Universal Adversarial Suffixes for Language Models Using Reinforcement Learning with Calibrated Reward 提出基于强化学习的通用对抗后缀方法，提升语言模型攻击的迁移性和有效性。 reinforcement learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
10 Ask, Answer, and Detect: Role-Playing LLMs for Personality Detection with Question-Conditioned Mixture-of-Experts 提出ROME框架，利用角色扮演LLM和混合专家模型进行人格检测。 semantic mapping semantic map large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
11 Curriculum Guided Massive Multi Agent System Solving For Robust Long Horizon Tasks 提出课程引导的大规模多智能体系统，解决鲁棒长时程任务 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页