cs.CL（2026-10-26）

📊 共 23 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 DynEdit: Dynamic Entropy-Guided Sequential Editing for Large Language Models 提出DynEdit以解决大语言模型知识编辑的挑战 large language model
2 When the Persona Contradicts the Line: Diagnosing Plan-to-Synthesis Failures in Chain-of-Thought TTS 提出双语基准以诊断链式思维TTS中的计划与合成失败 chain-of-thought
3 From Self-Confidence to Statistical Accuracy: Scaling Human-in-the-Loop Evaluation with Direct and LLM-Assisted Auditing 提出统一统计框架以提升人机协作评估效率 large language model multimodal
4 ImgMT: Wikipedia-based Dataset For Evaluating Text-in-Image Translation 提出ImgMT数据集以评估图像中的文本翻译 large language model multimodal
5 Beyond Words Alone: Disentangling Distributed Evidence for Visually Grounded Cultural Translation 提出Culture-MMT以解决社交媒体翻译中的文化适应问题 visual grounding
6 Adapting LLM-SC to Living Sacred Heritage: Domain-Specific Extraction, Semantic Networks and Human Evaluation on Pashupatinath Temple Reviews 提出LLM-SC框架以解决活态文化遗产数据提取问题 large language model
7 Constrained Decoding: Harmful or Beneficial? Revisiting Its Impact on LLM Performance 提出约束解码以提升大型语言模型的推理性能 large language model
8 Understanding Jev-Style System One Models: Does Structure Reduce Selection Bias? 探讨Jev风格系统一模型的结构是否减少选择偏差 large language model
9 Quantifying Claim Smoothing by LLM Research Agents 提出SmoothScore以量化LLM研究代理的主张平滑问题 large language model
10 Probing the phon* representations of the Mimi codebooks 探讨Mimi编码器的音素表示以优化语音编码 large language model
11 CRIM: Composable and Reusable Item Memory for Efficient Personalized Retrieval 提出CRIM框架以解决个性化检索中的高成本问题 large language model
12 Style over Substance? Linguistic Bias in LLM-based Peer Review Copy title 提出实验框架研究LLM在同行评审中的语言偏见问题 large language model
13 High Scores, Hidden Failures: Evaluating Codebook-Based LLM Content Analysis 评估LLM内容分析中的隐性失败与高分表现 large language model
14 Rethinking Evaluation for Principled Generalization Using a Kripkean Challenge to Meaning as Use 提出基于Kripke挑战的评估方法以解决语言模型泛化问题 large language model
15 FATE-5: Faithful Alignment of Trait Expression for Continuous Big Five Personality Control 提出FATE-5以解决个性控制与对话自然性之间的矛盾 large language model
16 RubriX: A Fast, Fine-Tunable Non-Autoregressive Rubric Judge 提出RubriX以解决基于评分标准的评估效率问题 large language model
17 Text Is the Bottleneck: A Small-Scale Replication of LLM Game Playing in Riichi Mahjong 提出小规模模型以解决大型语言模型在麻将游戏中的瓶颈问题 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🔗⭐
18 CalibraRL: Confidence-Calibrated Reinforcement Learning for Mitigating Hallucinations in RAG 提出CalibraRL以解决RAG中的幻觉问题 reinforcement learning reward design large language model
19 Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models 提出dSpark框架以加速扩散语言模型的生成效率 distillation large language model ✅
20 SimPOTrans: Self-Adaptation for Offline-to-Simultaneous Speech Translation via Reinforcement Learning 提出SimPOTrans以解决离线到同步语音翻译自适应问题 reinforcement learning large language model
21 DTCA: Denoising Token Credit Assignment for Reinforcement Learning in Diffusion Language Models 提出DTCA以解决扩散语言模型中的信用分配问题 reinforcement learning

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

#题目一句话要点标签🔗⭐
22 The Geometry of Emotion Representations Across Scale, Depth, and Instruction Tuning 研究情感几何在不同规模与调优下的表现 motion representation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
23 Evaluating DeepSeek V4.1 Flash-1017 for loss of control and harmful manipulation risks 评估DeepSeek V4.1 Flash-1017以应对控制丧失和有害操控风险 manipulation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页