cs.CL（2026-10-06）

📊 共 25 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (21 🔗4) 支柱一：机器人控制 (Robot Control) (2) 支柱二：RL算法与架构 (RL & Architecture) (2)

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

#题目一句话要点标签🔗⭐
1 Visual Abstention in Unified Multimodal Models 提出视觉放弃机制以提升多模态模型的生成能力 multimodal
2 Dynamic Positional Attention Modulation for Parameter-Efficient Fine-Tuning of Large Language Models 提出DyPAM以解决大语言模型微调中的位置编码问题 large language model
3 Towards In-Parameter Memory Augmentation for Large Language Models 提出基于参数内存增强的大型语言模型以解决知识更新问题 large language model
4 Incidental information contaminates patient notes and disrupts clinical reasoning in large language models 提出双重编码假设以解决临床推理中的信息污染问题 large language model
5 Pseudowords as probes: Large Language Models show little of the sublexical sensitivity that governs human pseudoword processing 探讨大型语言模型对伪词处理的敏感性不足 large language model
6 Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models 提出路由日志检测器以提升多模态安全性 multimodal
7 Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference 提出HARP框架以解决异质偏好下的LLM协调问题 large language model
8 HouseholdBench: Evaluating Large Language Models as Predictors of Household Economic Behavior 提出HouseholdBench以评估大型语言模型在家庭经济行为预测中的应用 large language model ✅
9 Two Vectors Replace In-Context Demos: Structured Task Adaptation via Embeddings 提出STAVE以解决大规模多模态模型的任务适应问题 large language model multimodal
10 Latent space bias directions in LLMs capture confidence, not fairness 分析大语言模型中的潜在空间偏差方向以提升公平性 large language model
11 Agentic AutoRAG: RAG Pipeline Optimization through Reasoning-Driven Agents 提出Agentic AutoRAG以优化RAG管道配置问题 large language model
12 Knowing When Not to Answer: Cross-Domain and Multi-Turn Generalization of Latent Underspecification Signals 提出跨领域和多轮对话的潜在未指定信号识别方法 large language model
13 Rethinking Faithfulness in LLMs: A Pairwise Context-Sensitive Perspective 提出PFaithBench以解决大语言模型的信度评估问题 large language model ✅
14 APEX: Speculate smarter, not deeper 提出APEX以优化语言模型推理速度与计算效率 large language model
15 No Transformer Beats Six Covariates: Long-Horizon Prediction of Depressive Symptoms from Childhood Essays 提出基于儿童作文的抑郁症预测模型，超越传统方法 large language model
16 The Missing Minimal Pair: Stereotype Evaluation in LLMs 提出双重最小对比以解决大语言模型中的偏见评估问题 large language model ✅
17 Language Unalignability: Why Some Concepts Resist Cross-Cultural Benchmark Evaluation 提出语言不对齐性理论以解决跨文化评估问题 large language model
18 Are Language Models Script-Aware? 探讨语言模型的脚本知识以解决生成语言不一致问题 large language model
19 ARIA: Audio-Driven Melody-Tone Relation Modeling for Cantonese Lyric Authoring 提出ARIA框架以解决粤语歌词创作中的音调与旋律关系建模问题 TAMP
20 Does Steering Break Your Model? A Multi-Dimensional Evaluation Suite for LLM Steering Methods 提出SteerScope以系统评估LLM激活引导方法的有效性与副作用 large language model
21 DLoop: Looped Speculative Decoding 提出DLoop以提升自回归生成的解码效率 large language model ✅

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
22 AdvSim2Real : Training Web Agents Against Adaptive Prompt Injection in a Web World Model 提出AdvSim2Real以解决Web代理在动态注入攻击下的鲁棒性问题 sim2real world model world models
23 Detecting LLM-Assisted Vietnamese Writing via Keystrokes under Behavioral Manipulation 提出基于击键动态的越南语写作检测方法以应对行为操控问题 manipulation large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
24 Rethinking Cross-Tokenizer On-Policy Distillation: From Alignment Coverage to Supervision Reliability 提出跨标记器的在政策蒸馏方法以提高监督可靠性 distillation
25 IdeaAnchor: Teaching LLMs to Turn Literature into Research Ideas 提出IdeaAnchor以解决文献基础研究创意生成问题 reinforcement learning distillation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页