cs.CL（2025-07-09）

📊 共 33 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (25 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱一：机器人控制 (Robot Control) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (25 篇)

#题目一句话要点标签🔗⭐
1 Prompt Perturbations Reveal Human-Like Biases in Large Language Model Survey Responses 提示扰动揭示大型语言模型在调查问卷中类人的偏差 large language model
2 Elite Polarization in European Parliamentary Speeches: a Novel Measurement Approach Using Large Language Models 利用大型语言模型进行政治人物情感分析，提出一种测量精英极化的新方法 large language model
3 Expediting data extraction using a large language model (LLM) and scoping review protocol: a methodological study within a complex scoping review 利用大型语言模型和范围界定审查协议加速数据提取 large language model
4 Enhancing Food-Domain Question Answering with a Multimodal Knowledge Graph: Hybrid QA Generation and Diversity Analysis 提出融合多模态知识图谱的食物领域问答框架，提升生成质量与多样性 multimodal
5 Large Language Model for Extracting Complex Contract Information in Industrial Scenes 提出一种基于大语言模型的工业场景复杂合同信息抽取方法 large language model
6 Integrating External Tools with Large Language Models to Improve Accuracy 提出Athena框架，集成外部工具显著提升LLM在教育场景下的问题解答准确率 large language model
7 InvestAlign: Overcoming Data Scarcity in Aligning Large Language Models with Investor Decision-Making Processes under Herd Behavior InvestAlign：解决羊群效应下LLM在投资者决策对齐中的数据稀缺问题 large language model ✅
8 ixi-GEN: Efficient Industrial sLLMs through Domain Adaptive Continual Pretraining ixi-GEN：通过领域自适应持续预训练提升工业界小规模LLM的效率 large language model foundation model
9 CRISP: Complex Reasoning with Interpretable Step-based Plans CRISP：通过可解释的步骤计划进行复杂推理，提升数学推理和代码生成能力 large language model chain-of-thought
10 Frontier LLMs Still Struggle with Simple Reasoning Tasks 前沿大语言模型在简单推理任务上仍面临挑战 large language model
11 Multi-Agent Retrieval-Augmented Framework for Evidence-Based Counterspeech Against Health Misinformation 提出多智能体检索增强框架，用于生成针对健康虚假信息的循证反驳言论 large language model
12 SynthTextEval: Synthetic Text Data Generation and Evaluation for High-Stakes Domains SynthTextEval：面向高风险领域的合成文本生成与评估工具包 large language model
13 Planted in Pretraining, Swayed by Finetuning: A Case Study on the Origins of Cognitive Biases in LLMs 研究揭示：LLM认知偏差主要源于预训练，微调影响有限 large language model
14 Exploring LLMs for Predicting Tutor Strategy and Student Outcomes in Dialogues 探索LLM在对话中预测导师策略和学生表现的能力 large language model
15 MultiJustice: A Chinese Dataset for Multi-Party, Multi-Charge Legal Prediction 提出MultiJustice数据集，用于评估LLM在多被告、多罪名法律预测中的性能 large language model ✅
16 Developing and Maintaining an Open-Source Repository of AI Evaluations: Challenges and Insights 提出开放源代码AI评估库管理框架以应对评估挑战 large language model
17 Adaptive Termination for Multi-round Parallel Reasoning: An Universal Semantic Entropy-Guided Framework 提出基于语义熵引导的自适应终止框架，提升多轮并行推理效率。 large language model
18 RAG Safety: Exploring Knowledge Poisoning Attacks to Retrieval-Augmented Generation 针对知识图谱增强的RAG系统，提出一种隐蔽的知识投毒攻击方法。 large language model
19 Text to model via SysML: Automated generation of dynamical system computational models from unstructured natural language text via enhanced System Modeling Language diagrams 提出一种基于SysML的文本到模型自动生成方法，加速工程动力系统设计与部署。 large language model
20 AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research AblationBench：用于评估AI辅助消融实验规划的基准测试套件 chain-of-thought
21 Checklist Engineering Empowers Multilingual LLM Judges 提出基于清单工程的CE-Judge框架，赋能多语言LLM评估任务。 large language model
22 On the Effect of Uncertainty on Layer-wise Inference Dynamics 研究表明LLM的不确定性预测对层间推理动态影响较小，但模型能力可能改变这一现象。 large language model
23 A Mathematical Theory of Discursive Networks 构建话语网络数学模型，通过互审机制提升大型语言模型的信息可靠性 large language model
24 SpindleKV: A Novel KV Cache Reduction Method Balancing Both Shallow and Deep Layers SpindleKV：一种平衡浅层和深层的新型KV缓存缩减方法 large language model
25 On the Robustness of Verbal Confidence of LLMs in Adversarial Attacks 研究表明，针对LLM语言置信度的对抗攻击能显著降低其可靠性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
26 Pun Intended: Multi-Agent Translation of Wordplay with Contrastive Learning and Phonetic-Semantic Embeddings 提出结合对比学习与语音-语义嵌入的多智能体翻译框架，用于解决双关语跨语言翻译难题。 contrastive learning HuMoR large language model
27 SCoRE: Streamlined Corpus-based Relation Extraction using Multi-Label Contrastive Learning and Bayesian kNN SCoRE：利用多标签对比学习和贝叶斯kNN的精简型语料库关系抽取 contrastive learning large language model
28 Rethinking Verification for LLM Code Generation: From Generation to Testing 提出SAGA框架，提升LLM代码生成测试用例的覆盖率和质量，改进代码评估。 reinforcement learning large language model
29 FuDoBa: Fusing Document and Knowledge Graph-based Representations with Bayesian Optimisation FuDoBa：融合文档与知识图谱表征，通过贝叶斯优化提升领域文档分类。 representation learning large language model
30 Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation 提出SambaY：一种高效的Decoder-Hybrid-Decoder架构，用于长文本生成推理。 reinforcement learning SSM state space model ✅

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
31 UniConv: Unifying Retrieval and Response Generation for Large Language Models in Conversations UniConv统一检索与生成，提升大型语言模型在对话搜索中的性能 UniCon large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
32 VisualTrap: A Stealthy Backdoor Attack on GUI Agents via Visual Grounding Manipulation VisualTrap：一种针对GUI智能体的隐蔽后门攻击，通过视觉定位操纵实现 manipulation visual grounding

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

#题目一句话要点标签🔗⭐
33 A Language-Driven Framework for Improving Personalized Recommendations: Merging LLMs with Traditional Algorithms 提出融合LLM的个性化推荐框架，提升传统算法对文本偏好的理解 HuMoR large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页