cs.CL（2026-08-03）

📊 共 22 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 PICTURE: Enhancing Theory-of-Mind in Large Language Models by Revealing, Not Hiding, Characters' Lack of Knowledge 提出PICTURE以解决大型语言模型的心智理论问题 large language model chain-of-thought
2 TRAM: Enhancing Multimodal Reasoning with Trajectory-Derived Auxiliary Memory 提出TRAM以增强多模态推理能力，解决信息遗失问题 multimodal visual grounding
3 Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation 提出稀疏注意力选择性评估框架以优化长上下文模型 foundation model
4 Divergent large language model predictions from convergent representations in ambiguous word pairs 通过层级分析揭示解码器模型处理词义歧义的机制 large language model
5 TELLER: Non-intrusive Cross-Layer Root-Cause Analysis for LLM Inference 提出TELLER框架以解决LLM推理中的根因分析问题 large language model multimodal
6 Cultural Awareness is Represented but Not Decoded: Tracing Mythological Knowledge across 18 Open-Source LLMs 提出跨文化知识追踪方法以解决LLM文化偏见问题 language conditioned
7 Can AI Agents Simulate A/B Test Outcomes? A Validation Framework for Agentic Experimentation 提出模拟随机对照试验框架以优化A/B测试结果 foundation model
8 TIDES: A Longitudinal Bilingual Dataset for Modeling Multi-Party Social Dynamics 提出TIDES数据集以解决多方社交动态建模问题 large language model
9 Training-Free versus Training-Based Intent Classification in LLMs: Accuracy, Robustness, and Failure Modes 比较训练无关与训练相关的意图分类方法以提升分类准确性 large language model
10 PredAct-Bench: Benchmarking Tool-Augmented Dialogue under Controlled Tool Noise 提出PREDACTBENCH以解决工具噪声下的对话系统评估问题 large language model
11 Fast and Accurate Quotation Attribution in Literary Texts 提出高效的引文归属方法以解决文学文本中的归属挑战 large language model ✅
12 CRISP: Critical Step Perception for Training Efficient Deep Search Agents 提出CRISP框架以提高深度搜索代理的训练效率 large language model
13 Human-LLM Alignment in Language Attitudes Toward Non-Native Japanese 研究人类与大型语言模型在日语非母语态度的对齐 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
14 Self-Improving Large Language Models via Progressive Experience Evolution 提出SPEE框架以解决自我提升语言模型的经验转化问题 reinforcement learning distillation large language model ✅
15 Douyin Multimodal Embedding Model Technical Report 提出Douyin多模态嵌入模型以解决高效检索与细粒度匹配问题 representation learning multimodal
16 Disentangled Contrastive Learning for Zero-Shot Multilingual Dense Retrieval 提出解耦对比学习以解决零样本多语言密集检索问题 representation learning contrastive learning zero-shot transfer
17 IACM-RL: Intent-Aware Context Management and Reinforcement Learning for Complex Tool Invocation under Dynamic Intent Fluctuations 提出IACM-RL以解决动态用户意图波动下的工具调用问题 reinforcement learning distillation
18 CAVE: Competence-Aware Visual Boundary Evidence Alignment for Video Temporal Grounding 提出CAVE以解决视频时间定位中的视觉证据对齐问题 reinforcement learning TAMP
19 Look Ahead Before You Distill: Future Trajectory Validation of Teacher Guidance for Agentic On-Policy Distillation 提出FutureBridge-OPD以解决多回合任务中的教师指导有效性问题 distillation ✅
20 Learning What to Remember: Test-Time Training via Context Distillation 提出测试时间上下文蒸馏方法以优化长上下文建模 distillation
21 Bole: Efficient Tree Speculation for Hybrid-Attention Language Models 提出Bole以解决混合注意力语言模型的树状推测效率问题 linear attention large language model
22 Cross-Domain Hybrid OPD for Generalizable Search Agents 提出跨域混合OPD框架以提升搜索代理的通用性 reinforcement learning distillation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页