cs.CL（2026-06-25）

📊 共 29 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) (22 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1)

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

#题目一句话要点标签🔗⭐
1 SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context 提出SocialPersona以解决个性化语言助手的多模态偏好推断问题 large language model multimodal TAMP
2 NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models 提出NuclearQAv2以评估大型语言模型在核工程领域的能力 large language model
3 Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions 提出审计框架敏感行为不稳定性以改善心理健康交互 large language model
4 \textsc{DiARC}: Distinguishing Positive and Negative Samples Helps Improving ARC-like Reasoning Ability of Large Language Models 提出DiARC以改善大型语言模型的ARC类推理能力 large language model ✅
5 The Riddle Riddle: Testing Flexible Reasoning in Large Language Models and Humans 提出谜题谜题范式以测试大型语言模型的灵活推理能力 large language model
6 FBK's Long-form SpeechLLMs for IWSLT 2026 Instruction Following 提出FBK的长形式SpeechLLMs以解决指令跟随问题 instruction following
7 Comparing BERT Sentence-Pair Classification and Few-Shot LLM Prompting for Detecting Threat and Solution Framing in German Climate News 比较BERT与少量示例LLM提示在德语气候新闻中的应用 large language model chain-of-thought
8 Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning 提出PEEU方法以解决小型MLLM任务规划能力不足问题 multimodal
9 ReaORE: Reasoning-Guided Progressive Open Relation Extraction Empowered by Large Reasoning Models 提出ReaORE以解决开放关系抽取中的泛化问题 large language model
10 Information-Aware KV Cache Compression for Long Reasoning 提出InfoKV以解决长推理中的KV缓存压缩问题 large language model
11 Beyond Logical Forms: LLM-Extracted Patterns for Fallacy Classification 提出基于LLM的模式提取方法以提升谬误分类效果 large language model
12 Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting 提出TempoWave以解决LLMs在时间序列预测中的数值嵌入问题 large language model ✅
13 Mapping Political-Elite Networks in Europe with a Multilingual Joint Entity-Relation Extraction Pipeline 提出多语言联合实体关系提取管道以解决政治精英网络映射问题 large language model
14 How Surprising Is Historical Italian to Language Models? Tokenization Tax, Comprehension Tax, and a Simple Mitigation 提出诊断框架以解决历史意大利语对语言模型的挑战 large language model
15 LMs as Task-Specific Knowledge Bases: An Interpretability Analysis 探讨语言模型作为任务特定知识库的可解释性分析 chain-of-thought
16 Adversarial Diffusion Across Modalities: A Fusion Survey of Attacks, Defenses, and Evaluation for Text, Vision, and Vision-Language Models 提出对抗性扩散方法以整合文本、视觉和视觉-语言模型的评估 large language model
17 Utilizing Cognitive Signals Generated during Human Reading to Enhance Keyphrase Extraction from Microblogs 利用认知信号提升微博客中的关键词提取效果 multimodal
18 Extracting Problem and Method Sentence from Scientific Papers: A Context-enhanced Transformer Using Formulaic Expression Desensitization 提出基于公式表达去敏感化的上下文增强变换器以解决科学论文中的问题与方法提取问题 large language model
19 Soft Token Alignment for Cross-Lingual Reasoning 提出SOLAR以解决跨语言推理中的软标记对齐问题 large language model
20 DiARC: Distinguishing Positive and Negative Samples Helps Improving ARC-like Reasoning Ability of Large Language Models 提出DiARC以改善大型语言模型的ARC类推理能力 large language model ✅
21 Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement 提出延迟验证机制以解决多智能体LLM信念不稳定问题 large language model
22 Cluster, Route, Escalate: Cascaded Framework for Cost-Aware LLM Serving 提出级联框架以降低大语言模型服务成本 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
23 Zero-shot Tweet-Level Stance Detection Enhanced by External Knowledge and Reflective Chain-of-Thought Reasoning 提出KIRP框架以解决零样本推文立场检测问题 contrastive learning chain-of-thought
24 OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning 提出OPID框架以解决稀疏奖励下的强化学习问题 reinforcement learning distillation ✅
25 Paved with True Intents: Intent-Aware Training Improves LLM Safety Classification Across Training Regimes 提出意图感知训练以提升大型语言模型的安全分类能力 reinforcement learning preference learning DPO
26 CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention 提出CARVE以解决记忆盲区问题和参数浪费 linear attention
27 Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization 提出Psy-CoT和RAPO以提升角色扮演智能体的表现 reinforcement learning chain-of-thought
28 Erase-then-Delta Attention: Decoupling Erase and Write Addresses in Delta-Rule Linear Attention 提出Erase-then-Delta Attention以解决记忆更新中的信息过时问题 linear attention

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

#题目一句话要点标签🔗⭐
29 Where Do Models Find Happiness? Emotion Vectors in Open-Source LLMs 探讨开源大语言模型中的情感向量及其表现 motion representation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页