cs.CL（2026-06-10）

📊 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) (14 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application 系统研究大语言模型的代理环境工程以推动模型能力演进 large language model
2 Adaptive Multi-Resolution Procedural Knowledge Compression for Large Language Models 提出SKIM框架以解决大语言模型中的技能压缩问题 large language model ✅
3 Detecting Sensitive Personal Information in Japanese Pre-Training Corpora for Large Language Models 构建SCPI数据集以检测日语文本中的敏感个人信息 large language model
4 Can AI Reason Like an Urban Planner? Benchmarking Large Language Models Against Professional Judgment 提出UPBench框架以评估AI在城市规划中的推理能力 large language model
5 ALIGNBEAM : Inference-Time Alignment Transfer via Cross-Vocabulary Logit Mixing 提出ALIGNBEAM以解决跨词汇模型安全性问题 large language model
6 Teaching Diffusion to Speculate Left-to-Right 提出三种干预措施以优化扩散语言模型的左到右推理 large language model
7 System Report for CCL25-Eval Task 5: New Dataset and LoRA-Fine-Tuned Qwen2.5 提出PoetryQwen以解决古典诗歌翻译与情感理解问题 large language model
8 Soft-Prompt Tuning for Fair and Efficient LLM Benchmark Evaluation 提出软提示调优以解决LLM基准评估不公问题 large language model
9 Automated Creativity Evaluation of Language Models Across Open-Ended Tasks 提出自动化框架以评估语言模型的创造力 large language model
10 Beyond Third-Person Audits: Situated Interaction Auditing for User-Centered LLM Bias Research 提出情境交互审计框架以解决LLM偏见研究盲点 large language model
11 On The Effectiveness-Fluency Trade-Off In LLM Conditioning: A Systematic Study 系统研究大语言模型条件化中的有效性与流畅性权衡 large language model
12 Dummy Backdoor as a Defense: Removing Unknown Backdoors via Shared Internal Mechanisms for Generative LLMs 提出共享内部机制的虚拟后门以解决未知后门问题 large language model
13 Improving Cross-Format Robustness in Language Models with Multi-Format Training 提出多格式训练以提升语言模型的跨格式鲁棒性 large language model
14 Multi-Agent Reasoning with Adaptive Worker Allocation for Stance Detection 提出多代理推理框架以解决立场检测中的聚合问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
15 Decoding Multimodal Cues: Unveiling the Implicit Meaning Behind Hateful Videos 提出IARE框架以实现可解释的仇恨视频检测 direct preference optimization multimodal chain-of-thought
16 GraspLLM: Towards Zero-Shot Generalization on Text-Attributed Graphs with LLMs 提出GraspLLM以解决文本属性图的零-shot泛化问题 contrastive learning large language model ✅
17 Hey Chat, Can You Teach Me? Structuring Socratic Dialogue for Human Learning in the Wild 提出知识图谱与对话策略结合以优化学习效果 PPO large language model
18 Doc-to-Atom: Learning to Compile and Compose Memory Atoms 提出Doc-to-Atom以解决长文档推理中的内存和效率问题 distillation large language model
19 Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization 提出RACES框架以解决强化学习环境构建效率问题 reinforcement learning large language model
20 Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Models 提出AGDO框架以解决扩散语言模型的随机掩蔽问题 reinforcement 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🔗⭐
21 When Does Language Matter? Multilingual Instructions Reveal Step-wise Language Sensitivity in Vision-Language-Action Models 提出逐步语言敏感性干预以提升VLA模型的语言鲁棒性 manipulation vision-language-action VLA

⬅️ 返回 cs.CL 首页 · 🏠 返回主页