cs.CL（2026-10-19）

📊 共 27 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) (21 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (21 篇)

#题目一句话要点标签🔗⭐
1 OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory 提出OpenVLHarness以解决多模态推理中的工具使用问题 large language model multimodal
2 TempoMoE: Identifying and Steering Fast-Slow Thinking Experts for Efficient Multimodal Reasoning 提出TempoMoE以解决多模态推理中的思维深度控制问题 multimodal chain-of-thought
3 Measurement Validity and Representation Dynamics of Confidence Signals in Large Language Models 比较三种信心信号测量方法以提高语言模型的可信度 large language model
4 Cognitive Map in Large Language Models 提出认知地图模型以增强大语言模型的推理能力 large language model
5 LatComm: Reusable Context Compression for Latent Communication between Large Language Models 提出LatComm以解决大语言模型间的上下文传递效率问题 large language model
6 No Matter Who You Are: Addressivity and Sycophancy in Large Language Models 提出地址性与谄媚行为的关系分析以解决LLM的谄媚问题 large language model
7 FisherEdit: Fisher Information Guided Knowledge Editing for Large Language Models 提出FisherEdit以解决大型语言模型知识编辑中的能力退化问题 large language model
8 MapSteal: Map-based Autonomous Knowledge Stealing Attack Against Large Language Models 提出MapSteal以解决自主知识获取挑战 large language model ✅
9 HoloCosm: Beyond Single-Metric Benchmark Compression Across Multiple Metrics 提出HoloCosm以解决多指标基准压缩问题 VLA multimodal
10 RefCraft: Benchmarking Visual-Reference-Conditioned Agentic Generation through Web Development 提出RefCraft基准以解决视觉参考条件下的生成挑战 multimodal
11 MemT-1.0: Towards Memory-Native Evolution 提出MemT-1.0以实现记忆本土化演化 foundation model
12 AdaExploreBench: Benchmarking LLM Agents for Proactive Exploration and Test-time Adaptation in Evolving Environments 提出AdaExploreBench以解决LLM代理在动态环境中的探索与适应问题 large language model
13 ASTOR: Benchmarking and Improving Autoformulation for Soft-Constraint Optimization with a Formal Grammar 提出ASTOR框架以改进软约束优化问题的自动公式化 large language model
14 EpiLens: Lightweight RAG for Epidemiological Knowledge Acquisition 提出EpiLens以解决流行病学知识获取问题 large language model
15 AfriGuard: Safety and Robust Refusal For African Languages 提出AfriGuard以解决非洲语言安全响应问题 large language model
16 Answering While You Type: Optimizing Time-to-Answer with Streaming Generation 提出流式生成以优化用户响应时间 large language model
17 Balance-Aware Pivot Selection for Parallel Diffusion LLM Decoding 提出平衡感知的枢轴选择以优化并行扩散语言模型解码 instruction following
18 Scaling Synthetic Pre-training Data and Evaluation for Indic LLMs 提出Nemotron-Pretraining-Indic-v1以提升印度语言模型的能力 instruction following
19 CALM: Consistency-Aware Logit Masking in Decoding Space for Mitigating Overthinking of Large Reasoning Models 提出CALM以解决大型推理模型过度思考问题 chain-of-thought
20 Truncated Reasoning Under a Budget: Evaluating System-1 Models as Decision-makers 提出Jev模型以优化有限推理预算下的决策过程 large language model
21 Inside the Scorer: Mechanistic Evidence on Rubric Control in LLM-Based Automated Essay Scoring from 25 Base Models 揭示LLM自动评分中的评分标准控制机制 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
22 Bridged Self-Distillation for On-Policy Learning 提出桥接自蒸馏方法以解决在线学习中的教师模型不匹配问题 policy learning distillation ✅
23 MITRA-lexicon: Lexica of attested translations from Sanskrit into English, Tibetan and Chinese 提出MITRA-lexicon以解决梵文翻译资源不足问题 teacher-student large language model
24 Learning to Reason Efficiently with Interventional Step Credit 提出干预步骤信用机制以提升推理效率 reinforcement learning chain-of-thought
25 SpecMRL: Redesign Matryoshka Representation Learning within Parametric Spectrum 提出SpecMRL以优化Matryoshka表示学习的特征空间 representation learning
26 Hierarchical Token Merging with Recursive Autoencoders for Unsupervised Grammar Induction 提出层次化标记合并的递归自编码器以解决无监督语法归纳问题 representation learning distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
27 Where Does Rhetorical Distortion Live? A Multi-Platform, Multi-Language Measurement of Influencer Content 提出多维度框架以测量网络影响者内容中的修辞扭曲 manipulation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页