cs.CL（2026-08-04）

📊 共 33 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) (23) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱七：动作重定向 (Motion Retargeting) (1)

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

#题目一句话要点标签🔗⭐
1 M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models 提出M-GATE基准以评估多语言模型的语法和翻译能力 large language model
2 VIBE: A VAD-Informed Benchmark for Entity-Centered Affective Profiling of Large Language Model Outputs 提出VIBE基准以解决大语言模型输出的情感分析问题 large language model
3 GPTKB 2.0: Direct Construction of Disambiguated Knowledge Bases from Large Language Models 提出GPTKB 2.0以解决知识库构建中的歧义问题 large language model
4 Cross-Lingual Bias in Large Language Models: A Comparative Analysis of English and Swahili 分析大型语言模型中的跨语言偏见以提升多语言安全性 large language model
5 DUD: Decoupled Update Dynamics for Reliable Uncertainty Quantification in Large Language Models 提出DUD框架以解决大语言模型的不确定性量化问题 large language model
6 On the Diversity of Analogy Making in Large Language Models 评估大型语言模型类比生成的多样性以促进创新 large language model
7 Scalable Frequency- and Length-Aware Subdocument Deduplication for Large Language Model Pretraining 提出可扩展的子文档去重框架以解决大规模预训练中的冗余问题 large language model
8 Activation-Guided Neuron Intervention to Induce Alzheimer's-Related Computational Language Phenotypes in a Large Language Model 提出激活引导神经元干预以揭示阿尔茨海默病相关语言特征 large language model
9 Beyond Accuracy: A Multidimensional Evaluation of Statistical Reasoning in Large Language Models 提出多维评估框架以提升大型语言模型的统计推理能力 large language model
10 Sensitivity, Causality, and Repair Dissociate: A Layer-Wise Analysis of Perturbation Robustness and Its Scaling 提出层级分析方法以解决语言模型对扰动输入的鲁棒性问题 chain-of-thought
11 Beyond Initialization Loss: A Systematic Study of Token Embedding Initialization Strategies for LLM Vocabulary Extension 提出多种初始化策略以优化大语言模型词汇扩展 large language model
12 MultiGlobeQA: A Multilingual and Globally Diverse Benchmark for Geospatial Reasoning 提出MultiGlobeQA以解决地理空间推理的多语言基准问题 large language model
13 Beyond Representational Similarity: Source-Conditioned Description-Length Gain for Generative Plagiarism Detection and Candidate Source Reranking 提出源条件描述长度增益以解决生成性抄袭检测问题 large language model
14 How Closely Do LLM Reviews Align with Human Peer Review? 比较大型语言模型与人类评审的一致性 large language model
15 Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation 提出区分模拟与估计以优化语言模型的人类意见模拟 large language model
16 SocietyBench: Forecasting Counterfactual Social-World Evolution 提出SocietyBench以评估社会事件预测能力 large language model
17 WorldCup Arena: Prospective, Leakage-Free Evaluation of Frontier LLMs on a Live Tournament 提出一种无泄漏的前沿LLM评估方法以预测世界杯赛事 large language model
18 ConlangBench: Exploring Language Knowledge and Learning in LLMs through Diverse Constructed Languages 提出ConlangBench以评估大语言模型在构造语言学习中的表现 large language model
19 Don't Let Me Ask for It: LLMs Show Deficiencies in Active Multi-Turn Information Acquisition for Abductive Inference 提出Alien Abduction游戏以研究LLMs在推理中的信息获取能力 large language model
20 Relational Priors as Convergence Pressure in LLM-Based Multi-Agent Systems 提出关系先验作为LLM-MAS中的收敛压力以优化多智能体系统 large language model
21 ICO: Enhancing Semantic-Shift Jailbreaks via Iterative Context Optimization 提出ICO框架以优化语义转移越狱攻击效果 foundation model
22 What Language Does and What the Evidence Supports: A Functional Role Taxonomy and Evidence Audit of Language Grounding in Embodied Agents 提出语言功能角色分类以评估语言在具身智能体中的作用 foundation model
23 SeqLLM: Augmenting LLMs with Behavioral-Sequence Modeling for High-Stakes Decisions at WeChat Pay 提出SeqLLM以解决微信支付商户风险控制问题 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🔗⭐
24 CVPO: Enhancing LLM Reinforcement Learning Reasoning via Value-Variance Adaptation and Dynamic Curriculum Learning 提出CVPO以解决大语言模型推理能力不足的问题 reinforcement learning curriculum learning large language model
25 TQLite: Multi-LLM Jury Guided Distillation for Real-time MQM Translation Quality Evaluation 提出TQLite以解决小型语言模型在翻译质量评估中的性能不足问题 distillation large language model
26 Aligning Large Vision-Language Models at Test Time: A Trajectory-Guided Structured Sampling Approach 提出轨迹引导的结构化采样方法以解决视觉语言模型对齐问题 reinforcement learning multimodal visual grounding
27 Internalizing Academic Writing Workflows for Introduction Generation via Struct-Aware Policy Learning 提出StructPO框架以优化学术论文引言生成流程 policy learning large language model
28 Hi-TTRL: Regulating Consensus with Hints for Test-Time Reinforcement Learning 提出Hi-TTRL以解决测试时强化学习中的共识调节问题 reinforcement learning large language model
29 TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning 提出TurnSight以解决工具集成推理中的细粒度信用分配问题 reinforcement learning distillation ✅
30 Language-Specialized Multi-Teacher On-Policy Distillation for Multilingual LLM-Based ASR 提出语言专用多教师在线蒸馏以解决多语言ASR优化冲突问题 reinforcement learning distillation
31 Efficient Knowledge Distillation for LLMs: Offline Top-K Logits and a Fused Chunked KL Loss 提出高效知识蒸馏方法以提升小型语言模型训练效率 distillation ✅
32 SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs 提出Parallel-RL以解决多任务学习中的任务冲突问题 reinforcement learning large language model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
33 ArtECulture: Benchmarking Culture-Conditioned Visual Emotion Understanding in Multimodal Large Language Models 提出ArtECulture基准以解决文化条件下视觉情感理解问题 motion prediction large language model multimodal

⬅️ 返回 cs.CL 首页 · 🏠 返回主页