cs.CL（2026-07-20）

📊 共 41 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) (31 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱一：机器人控制 (Robot Control) (1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (31 篇)

#题目一句话要点标签🔗⭐
1 Large Language Models as Unified Multimodal Learners for Clinical Prediction 提出统一多模态学习模型以简化临床预测系统 large language model multimodal
2 Hidden in Thought: Transferable Chain-of-Thought Artifacts Induce Harmful Behavior 提出可转移的思维链条以应对语言模型的有害行为 chain-of-thought
3 AuAu: A Benchmark for Auditing Authoritarian Alignment in Large Language Models 提出AuAu基准以审计大型语言模型中的威权主义倾向 large language model
4 An MLIR-Based Compilation Method for Large Language Models 提出基于MLIR的编译方法以解决大语言模型部署问题 large language model
5 PolyInterview: An LLM-based Platform for Immersive Mock Interview Practice with Comprehensive Multimodal Assessment 提出PolyInterview以解决面试准备中的实践不足问题 multimodal
6 Verbalizable Representations Form a Global Workspace in Language Models 提出Jacobian透镜技术揭示语言模型的全球工作空间特征 large language model
7 VarRate: Training-Free Variable-Rate KV Cache Compression for Long-Context LLMs 提出VarRate以解决长上下文LLM推理中的KV缓存瓶颈问题 large language model
8 DECODEM: Data Extraction from Corporate Organizational Documents via Enhanced Methods 提出DECODEM以解决企业组织文件数据提取问题 large language model
9 Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability 通过机制可解释性分析细调LLM中的道德偏见 large language model
10 Scaling Point-in-Time Language Models 提出点时语言模型以解决未来信息泄露问题 large language model
11 What Models Express, Suppress, and Resist: Auditing Open-Weight LLMs with Persona Vectors 提出人格向量审计开放权重LLM以揭示模型行为 chain-of-thought
12 EpiNarrate: Agentic Generation of Grounded Narratives from Epidemiological Scenario Projections 提出EpiNarrate以解决公共卫生叙事生成问题 large language model
13 From Plausible to Actionable: A Position on LLM Self-Explanations 提出自我解释的可行性评估框架以提升LLM的可解释性 large language model
14 BayesPO: Bayesian Prompt Optimization via Parallel-Tempered Gradient-Guided Discrete MCMC 提出BayesPO以优化大语言模型的提示生成 large language model
15 Frontier Language Models Struggle to Copy: Text Can Be Better Viewed in 2D 提出2D-RoPE以解决语言模型复制问题 large language model
16 From Articles to Premises: Building PrimeFacts, an Extraction Methodology and Resource for Fact-Checking Evidence 提出PrimeFacts以解决事实核查证据提取问题 large language model
17 What Should a Skill Remember? Quality--Cost Trade-offs in Cost-Aware Skill Rewriting for Language Model Agents 提出成本意识的技能重写方法以优化语言模型代理的性能 large language model
18 It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO 提出单例GRPO以揭示大语言模型的偏见脆弱性 large language model
19 Pancasila-Dilemmas: Evaluating Large Language Models on Indonesian Human Value Dilemmas Grounded in Pancasila 提出Pancasila-Dilemmas评估数据集以解决印尼价值观对齐问题 large language model ✅
20 Large Language Models for Citation Function Classification 提出大型语言模型进行引文功能分类以提升文献分析 large language model
21 How Reliable Are Multimodal Signals of Conversational State? Evidence from Remote Dyadic Collaborative Tasks 提出三维评估框架以提升多模态对话状态信号的可靠性 multimodal
22 Benchmarking Resource-Efficient LLMs for Research Topic Ontology Generation in the Biomedical Field 提出小型LLM以解决生物医学领域本体生成问题 large language model chain-of-thought
23 PPL-Factory: Task-Aware and Budget-Aware Data Selection from Language Modeling to Reasoning 提出PPL-Factory以解决数据选择中的任务感知与预算感知问题 large language model
24 DeLIVeR: Decomposed Learning for Information-grounded Veracity Recognition via Reinforced Knowledge Graph Exploration 提出DeLIVeR以解决自动化事实核查中的查询脆弱性问题 large language model
25 Zero Hallucination, by Construction: Hallucination-Aware Layered Oversight for Trustworthy Enterprise AI 提出HALO框架以解决企业AI的幻觉问题 large language model
26 It's Not What You Say, It's How You Say It: Evaluating LLM Responses to Expressions of Belief 提出一种评估EoB对LLM响应影响的分类法 large language model
27 VDAR-Router: Adaptive LLMs Routing via Verbalized Query Difficulty Analysis Retrieval 提出VDAR-Router以解决LLM路由中的查询难度分析问题 large language model
28 When a Name Is Not a Name: A Benchmark Dataset and Distilled Reasoning for Culturally Entangled Bangla Homographs in Low-Resource LLMs 提出文化纠缠同形异义词消歧义方法以解决低资源语言模型的偏见问题 chain-of-thought ✅
29 ESCUCHA: A Spanish Speech Benchmark for Heterogeneous Acoustic Conditions 提出ESCUCHA基准以评估西班牙语在多样化声学条件下的理解能力 multimodal
30 C$^2$KV: Compressed and Composable KV Cache Reuse for Efficient LLM Inference 提出C$^2$KV以解决长上下文LLM推理中的KV缓存存储与访问问题 large language model
31 D-NOVA: In-Storage Retrieval Accelerator via Dual-Bound 3D NAND-Optimized Similarity Search with Vector Adaptation 提出D-NOVA以解决存储检索加速中的性能瓶颈问题 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🔗⭐
32 RLearner-LLM: Balancing Logical Grounding and Fluency in Large Language Models via Hybrid Direct Preference Optimization 提出RLearner-LLM以解决逻辑对齐与流畅性之间的平衡问题 PPO RLHF DPO
33 KDFlow: A User-Friendly and Efficient Knowledge Distillation Framework for Large Language Models 提出KDFlow以提升大语言模型蒸馏效率 distillation large language model
34 ToolSciVer: Multimodal Scientific Claim Verification with Visual Tool Augmented Reinforcement Learning 提出ToolSciVer以解决多模态科学声明验证问题 reinforcement learning multimodal
35 Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents 提出多代理框架以模拟和审计政治联盟形成 reinforcement learning RLHF DPO
36 Brain-CLIPLM: Semantic Compression for EEG-to-Text Decoding 提出Brain-CLIPLM以解决EEG到文本解码的语义压缩问题 contrastive learning large language model chain-of-thought
37 Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning 提出BIRD方法以解决推理模型冗余计算问题 distillation chain-of-thought
38 Decoupled Alignment for Robust Plug-and-Play Adaptation 提出无监督对齐方法以增强大语言模型的安全性 reinforcement learning distillation large language model
39 When Perplexity Lies: Generation-Focused Distillation of Hybrid Sequence Models 提出GenDistill以优化混合序列模型的生成性能 distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
40 Bifocal Attention: Harmonizing Geometric and Spectral Positional Embeddings for Algorithmic Generalization 提出双焦点注意力以解决算法推理中的结构差距问题 manipulation large language model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
41 Loop the Loopies! 提出Loopie以解决循环Transformer性能不足问题 IMoS

⬅️ 返回 cs.CL 首页 · 🏠 返回主页