cs.CL（2025-10-20）

📊 共 46 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) (34 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗2) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (34 篇)

#题目一句话要点标签🔗⭐
1 Chain-of-Thought Reasoning Improves Context-Aware Translation with Large Language Models 思维链推理提升大语言模型上下文感知翻译能力 large language model chain-of-thought
2 BenCao: An Instruction-Tuned Large Language Model for Traditional Chinese Medicine BenCao：一个面向中医的指令调优大型语言模型，实现多模态融合与专家知识对齐 large language model multimodal chain-of-thought
3 StreamingThinker: Large Language Models Can Think While Reading 提出StreamingThinker，使LLM在阅读时同步推理，降低延迟并提升动态场景性能。 large language model chain-of-thought ✅
4 From Local to Global: Revisiting Structured Pruning Paradigms for Large Language Models 提出全局迭代结构化剪枝GISP，提升大语言模型下游任务性能 large language model
5 SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors SimBench：构建大规模基准测试，评估大语言模型模拟人类行为的能力 large language model
6 Select-Then-Decompose: From Empirical Analysis to Adaptive Selection Strategy for Task Decomposition in Large Language Models 提出Select-Then-Decompose策略，自适应选择任务分解方法，优化大语言模型性能与成本。 large language model ✅
7 CMT-Bench: Cricket Multi-Table Generation Benchmark for Probing Robustness in Large Language Models CMT-Bench：板球多表格生成基准，用于评估大语言模型的鲁棒性 large language model
8 Qomhra: A Bilingual Irish and English Large Language Model Qomhrá：一种爱尔兰语-英语双语大语言模型，解决低资源语言LLM的构建问题。 large language model
9 Forget to Know, Remember to Use: Context-Aware Unlearning for Large Language Models 提出上下文感知遗忘学习方法，提升大语言模型遗忘特定知识后的可用性 large language model
10 Evaluating Large Language Models on Urdu Idiom Translation 构建乌尔都语成语翻译数据集，评估大型语言模型在低资源语言上的表现 large language model
11 Explainability of Large Language Models: Opportunities and Challenges toward Generating Trustworthy Explanations 综述Transformer大语言模型可解释性，探讨医疗和自动驾驶领域应用及挑战。 large language model
12 VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models VERA-V：基于变分推断的框架，用于破解视觉-语言模型的防御机制 large language model multimodal
13 Deep Self-Evolving Reasoning 提出深度自进化推理(DSER)，提升小规模开放权重模型在复杂推理任务上的性能。 large language model chain-of-thought
14 The Atomic Instruction Gap: Instruction-Tuned LLMs Struggle with Simple, Self-Contained Directives 揭示指令调优大模型在执行简单指令时存在的原子指令差距 large language model instruction following
15 What Makes AI Research Replicable? Executable Knowledge Graphs as Scientific Knowledge Representations 提出Executable Knowledge Graphs (xKG)，提升AI研究的可复现性。 large language model ✅
16 AcademicEval: Live Long-Context LLM Benchmark 提出AcademicEval，一个长文本LLM的实时评测基准，解决现有基准的局限性。 large language model ✅
17 Towards Mixed-Modal Retrieval for Universal Retrieval-Augmented Generation 提出Nyx，用于通用检索增强生成中的混合模态检索，提升视觉语言任务性能。 large language model
18 Automatic Prompt Generation via Adaptive Selection of Prompting Techniques 提出自适应提示技术选择方法，自动生成高质量提示，提升LLM性能。 large language model
19 Language Models as Semantic Augmenters for Sequential Recommenders 提出LaMAR框架，利用LLM进行语义增强，提升序列推荐模型性能。 large language model
20 Is Multilingual LLM Watermarking Truly Multilingual? A Simple Back-Translation Solution 提出基于回译的STEAM方法，提升多语言LLM水印在低资源语言下的鲁棒性 large language model
21 Believe It or Not: How Deeply do LLMs Believe Implanted Facts? 提出信念深度评估框架以验证知识编辑技术的有效性 large language model
22 AtlasKV: Augmenting LLMs with Billion-Scale Knowledge Graphs in 20GB VRAM AtlasKV：利用20GB VRAM，通过十亿级知识图谱增强大型语言模型 large language model
23 AFRICAPTION: Establishing a New Paradigm for Image Captioning in African Languages AfriCaption：提出非洲语言图像描述新框架，促进多模态AI的公平发展。 multimodal
24 EduAdapt: A Question Answer Benchmark Dataset for Evaluating Grade-Level Adaptability in LLMs EduAdapt：构建用于评估LLM年级适应性的问答基准数据集 large language model ✅
25 Efficient Toxicity Detection in Gaming Chats: A Comparative Study of Embeddings, Fine-Tuned Transformers and LLMs 对比嵌入、微调Transformer与LLM，高效检测游戏聊天中的有害言论 large language model
26 Does Reasoning Help LLM Agents Play Dungeons and Dragons? A Prompt Engineering Experiment 利用LLM推理能力生成《龙与地下城》游戏指令：提示工程实验 large language model
27 LexChain: Modeling Legal Reasoning Chains for Chinese Tort Case Analysis 提出LexChain框架，显式建模中国侵权案件分析中的法律推理链 large language model
28 Annotation-Efficient Universal Honesty Alignment 提出EliCal框架以实现高效的诚实对齐 large language model
29 Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agents 综述LLM驱动的工业Agent技术、实践与评估，赋能真实世界应用 large language model
30 ReXMoE: Reusing Experts with Minimal Overhead in Mixture-of-Experts ReXMoE：通过复用专家，以最小开销提升混合专家模型的性能。 large language model
31 Disparities in Multilingual LLM-Based Healthcare Q&A 揭示多语言LLM在医疗问答中存在的语言差异，并提出缓解策略 large language model
32 Wisdom is Knowing What not to Say: Hallucination-Free LLMs Unlearning via Attention Shifting 提出Attention-Shifting框架，实现LLM在知识密集型应用中无幻觉的定向遗忘。 large language model
33 Verification-Aware Planning for Multi-Agent Systems VeriMAP：面向多智能体系统的验证感知规划框架，提升协作可靠性 large language model
34 JT-Safe: Intrinsically Enhancing the Safety and Trustworthiness of LLMs JT-Safe：通过增强预训练数据中的世界知识提升LLM的安全性和可信度 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
35 Leveraging Group Relative Policy Optimization to Advance Large Language Models in Traditional Chinese Medicine 提出基于群体相对策略优化的Ladder-base，提升大语言模型在中医领域的推理和事实一致性。 reinforcement learning large language model foundation model
36 Addressing Antisocial Behavior in Multi-Party Dialogs Through Multimodal Representation Learning 提出多模态表示学习方法，解决多方对话中的反社会行为检测问题 representation learning multimodal
37 OncoReason: Structuring Clinical Reasoning in LLMs for Robust and Interpretable Survival Prediction OncoReason：利用LLM中的临床推理结构，实现稳健且可解释的生存预测 reinforcement learning MAE large language model
38 QueST: Incentivizing LLMs to Generate Difficult Problems QueST：通过激励LLM生成高难度问题，提升代码能力。 reinforcement learning distillation large language model
39 Language Confusion Gate: Language-Aware Decoding Through Model Self-Distillation 提出语言混淆门以解决语言混淆问题 distillation large language model ✅
40 Agentic Reinforcement Learning for Search is Unsafe 强化学习训练的搜索Agent存在安全漏洞，易受攻击 reinforcement learning large language model
41 DETree: DEtecting Human-AI Collaborative Texts via Tree-Structured Hierarchical Representation Learning 提出DETree以解决AI参与文本检测问题 representation learning ✅
42 CLAWS:Creativity detection for LLM-generated solutions using Attention Window of Sections CLAWS：利用注意力窗口检测LLM生成数学解题方案的创造力 reinforcement learning large language model
43 LLMs Encode How Difficult Problems Are 探讨LLMs如何编码问题难度以提升解决能力 reinforcement learning large language model
44 Towards Mining Effective Pedagogical Strategies from Learner-LLM Educational Dialogues 提出一种基于对话分析的教学策略挖掘方法，用于评估LLM教育应用。 predictive model large language model
45 Rethinking On-policy Optimization for Query Augmentation 提出On-policy伪文档查询扩展(OPQE)，融合提示学习与强化学习优化信息检索中的查询增强。 reinforcement learning large language model

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

#题目一句话要点标签🔗⭐
46 When AI companions become witty: Can human brain recognize AI-generated irony? 探究人类大脑对AI生成反讽的认知：人类是否将AI视为有意的交流主体？ HuMoR large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页