cs.CL（2026-09-29）

📊 共 33 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) (22 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗3) 支柱四：生成式动作 (Generative Motion) (1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (22 篇)

#题目一句话要点标签🔗⭐
1 Devils in Question Relay: Source-Conditioned Relay Steering to Mitigate Hallucinations in Audio-visual Large Language Models 提出SECRET以解决音视频大语言模型中的源混淆问题 large language model multimodal
2 RAEGNet: Relation-Aware Evidence Graph Network for Harm-Aware Multimodal Fake News Detection 提出RAEGNet以解决多模态假新闻检测中的危害感知问题 multimodal
3 Rethinking Multimodal Fake News Detection in the Generative AI Era 提出GAHR框架以解决生成内容时代的多模态假新闻检测问题 multimodal
4 Correct Answers, Invalid Traces: What Verifiable Grade-School Math Reveals About Chain-of-Thought Traces 提出iGSM基准以验证链式思维的有效性 chain-of-thought
5 A Proposed Rubric for Evaluating Expressed Clinical Reasoning in Large Language Model Responses 提出临床推理评估量表以提升语言模型响应质量 large language model
6 From Routing Signals to Selective Review: Visual regrounding in MoE VLMs 提出基于路由信号的选择性复审框架以解决目标缺失问题 visual grounding
7 How Many Labels Does a Language Need? Annotation Budgets and Cross-Lingual Pooling for African-Language Text Classification 提出跨语言标签池化方法以优化非洲语言文本分类 zero-shot transfer
8 Gender bias across LLMs is common and highly heterogenous 分析大型语言模型中的性别偏见及其异质性 large language model
9 The Unequal Influence of Bad Advice: Using Training Data Attribution to Modulate Emergent Misalignment 利用训练数据归因调节新兴失调现象的影响 large language model
10 Effective Dense Retrieval using Only In-Context Examples 提出RICE方法以实现无训练的高效密集检索 large language model ✅
11 AnthroDial: Benchmarking LLM Anthropomorphism in Autonomous Social Interaction 提出AnthroDial框架以提升自主社交智能体的人性化交互能力 large language model
12 Selecting What Matters: Semantic Compression-Guided Selective Pooling for Long-Context Embeddings 提出SCSP框架以解决长文档语义信息稀释问题 large language model
13 Reader Proficiency Shapes Layer-wise Surprisal Profiles 探讨读者熟练度对层级惊讶度特征的影响 large language model
14 Pair Difficulty Matters: Rethinking Pairwise LLM-as-a-Judge Evaluation and Consistency 提出基于排名差距的评估方法以改进LLM评估一致性 large language model ✅
15 MERGE: Multi-LLM Ensemble for Retrieval via Generative Enrichment 提出MERGE框架以提升信息检索中的查询扩展效果 large language model
16 RunyaNER: Auxiliary Language Selection for Runyankore NER 提出RunyaNER以解决Runyankore语言的命名实体识别问题 zero-shot transfer
17 CypherTurn: A Multi-Turn Benchmark for Conversational Text-to-Cypher Evaluation and the Autonomy Divergence 提出CypherTurn以解决多轮对话图数据库查询评估问题 instruction following ✅
18 AMU:Admission and Memory Update for Personalized Conversations---Structured Memory with SLM Guided Control 提出AMU框架以解决个性化对话中的记忆管理问题 large language model
19 ATTUNER: Recomputation-Free KV Cache Reuse via Query-Side Adaptation 提出ATTUNER以解决KV缓存重计算问题 large language model
20 SEED: Self-Speculative Decoding via Implicit Encoder-Decoder 提出SEED以解决大语言模型推理速度与质量的权衡问题 large language model ✅
21 When Updating Stops Being Learning: Rethinking LLM Self-Evolution via learnable information gain 提出基于可学习信息增益的自我进化框架以解决LLM性能退化问题 large language model
22 FinRT: Distilling Adaptive Red-Teaming Strategies into Reusable Adversarial Generators in Consumer Finance 提出FinRT以解决消费金融领域的对抗性生成问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
23 Distilling What Matters: Confidence-Aware Selective Distillation for Large Language Models 提出CaRE-KD以解决大语言模型蒸馏中的不确定性问题 distillation large language model instruction following
24 SRJudge: Empowering Large Language Models with Selective Reasoning for Fine-Grained Knowledge Concept Tagging 提出SRJudge框架以解决知识概念标记中的选择性推理问题 reinforcement learning large language model ✅
25 ER-JEPA: Experience Replay Improves Joint-Embedding Predictive Learning in Language Models 提出ER-JEPA以解决语言模型预测不稳定问题 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
26 Learning from Think-Mode Advantage via On-Policy Distillation 提出ThinkOPD以提升大语言模型的推理能力 teacher-student distillation large language model
27 VAA-CSEC: Vote-guided Advantage Allocation for Chinese Semantic Error Correction 提出VAA-CSEC以解决中文语义错误纠正问题 reinforcement learning distillation chain-of-thought
28 From Dissonance to Orchestration: Teacher Intervention in On-Policy Distillation 提出MAESTRO以优化教师干预在政策蒸馏中的应用 distillation ✅
29 VLM Fine-Tuning for End-to-End Combinatorial Optimization 提出视觉语言模型以优化组合优化问题 reinforcement learning large language model
30 STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization 提出STEPQuant以解决递归状态量化中的错误传播问题 dreamer linear attention ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
31 E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models 提出E-MoE以解决MDM生成质量不足问题 MDM

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

#题目一句话要点标签🔗⭐
32 BITEM at the NTCIR-19 R2C2 Task: Predicting Confidence from Agentic RAG Pipeline Signals 提出基于代理RAG管道信号的置信度预测方法 HMR

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
33 Video2Skill: From Streaming Experience to Reusable Embodied Skills 提出Video2Skill以解决技能重用与发现问题 manipulation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页