cs.CL（2026-09-03）

📊 共 28 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning 提出基于优势评估的推理步骤重要性分析方法 chain-of-thought
2 Knowledge Acquisition During Pre-training? Large Language Models Learn Better With Auxiliary Views 提出辅助视角以提升大型语言模型的知识获取能力 large language model
3 Investigating the Ability of Large Language Models to Analyze Recipes for Diabetes 研究大型语言模型分析糖尿病食谱的能力 large language model
4 IndicSafeEval: Safety Robustness of Large Language Models under Multilingual Persuasive Jailbreak Attacks 提出IndicSafeEval框架以评估多语言环境下LLM的安全性 large language model ✅
5 The Dice Roll Method: A Standardized Protocol for Repeated-Query Auditing of Large Language Model Brand Recommendations 提出骰子投掷法以解决大语言模型品牌推荐审计问题 large language model
6 Evaluating Criterion-Conditioned Behaviour of Large Language Models in Content Moderation 提出DECO方法以解决LLMs内容审核中的标准化问题 large language model
7 Typological Feature Prediction with Large Language Models: An In-Context Learning Approach 利用大型语言模型进行类型特征预测以解决可解释性不足问题 large language model
8 KhatianDoc: A Human-Verified Benchmark Diagnosing Multimodal LLM Failure on Bengali Legal Land Records 提出KhatianDoc基准以解决孟加拉法律土地记录的多模态LLM识别问题 multimodal
9 To What Extent Do Large Language Models Understand Bangla Idioms? 提出孟加拉语习语理解基准数据集以提升LLM性能 large language model
10 How Perturbations Propagate: A Multi-Level Analysis of Robustness in Large Language Models 提出多层次分析方法以评估大语言模型的鲁棒性 large language model
11 Flip, Don't Shuffle: Watermarking LLMs at the Speed of Inference 提出无状态伯努利水印以提升大语言模型的水印效率 large language model
12 What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation 提出成本有效的测试时计算方法以优化对话生成的修订传播 large language model ✅
13 Representational alignment yields generalizable safety in language models 提出表征对齐方法以提升语言模型的安全性 large language model
14 Beyond BLEU: A Case for Redefining Sign Language Translation Benchmarks 提出新评估方法以改进手语翻译基准 multimodal
15 </think> Doesn't Stop Reasoning: Analysis of Spurious CoT Termination 提出EoT注入策略以解决CoT推理终止问题 chain-of-thought ✅
16 Remember and Reweight: Enhancing Multi-Agent Debate with Experience Memory and Confidence Estimation 提出R²-MAD以解决多代理辩论中的共享误解问题 large language model
17 When Users Don't Ask: Benchmarking Context-Driven Memory Retrieval in Conversational Agents 提出LOCOMO-CONV基准以解决对话代理中的记忆检索问题 large language model
18 CROCODIL: Cross-Model Code Editing with LLMs 提出CROCODIL以解决跨模型代码编辑中的过度修改问题 large language model ✅
19 What Do CAE Simulation Agents Really Need Beyond a Generic Harness? 提出CAE仿真代理以优化仿真设置与执行效率 large language model
20 Lost in Reordering: Structural Sensitivity of Multilingual LLMs under Semantics-Preserving Perturbations 研究多语言LLM在语义保持扰动下的结构敏感性 large language model
21 Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning 提出随机注意力机制以解决KV缓存驱逐效率问题 large language model ✅
22 FrameBench:A Language Understanding Benchmark Based on Frame Semantics 提出FrameBench基准以解决语言理解中的隐性信息问题 large language model ✅
23 SGD-KV: Summarization Guided KV Cache Compression 提出SGD-KV以解决长上下文推理中的KV缓存压缩问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
24 Two-Stage Reinforcement Learning for Sound and Adversarial Test Generation in Code LLMs 提出两阶段强化学习框架以生成有效的代码测试用例 reinforcement learning large language model
25 Sequential Beats Joint: On the Interplay between On-Policy Distillation and RLVR 提出OPD-then-RL以优化强化学习与可验证奖励的结合 reinforcement learning distillation
26 ALRA: Adaptive Local Relational Alignment for Logit-Based Pre-training Distillation of Autoregressive Language Models 提出自适应局部关系对齐方法解决语言模型蒸馏问题 distillation
27 Decoupling Turn-Taking from Semantics: A Decoupled Data Approach for Finite-State-Machine-Based Full-Duplex Dialogue 提出解耦数据方法以提升全双工对话的自然性 SAC large language model ✅

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
28 PACE: Towards Surfacing Hidden Conflicts in User Requests 提出PACE以解决用户请求中的隐性冲突问题 egocentric

⬅️ 返回 cs.CL 首页 · 🏠 返回主页