cs.CL（2025-03-31）

📊 共 34 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (28 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (4) 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) (28 篇)

#题目一句话要点标签🔗⭐
1 AdaMMS: Model Merging for Heterogeneous Multimodal Large Language Models with Unsupervised Coefficient Optimization AdaMMS：面向异构多模态大语言模型的无监督系数优化模型融合 large language model multimodal
2 Model Hemorrhage and the Robustness Limits of Large Language Models 研究LLM的“模型出血”现象，提出缓解策略以提升模型部署时的鲁棒性。 large language model
3 Does "Reasoning" with Large Language Models Improve Recognizing, Generating, and Reframing Unhelpful Thoughts? 利用LLM推理能力改进认知行为疗法中不良想法的识别、生成与重构 large language model
4 Harnessing the Reasoning Economy: A Survey of Efficient Reasoning for Large Language Models 综述：探索大语言模型中的高效推理经济性 large language model
5 BEATS: Bias Evaluation and Assessment Test Suite for Large Language Models BEATS：用于评估大型语言模型偏见、伦理、公平性和事实性的测试套件 large language model
6 A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well? 对大语言模型测试时扩展（TTS）进行全面综述，分析其原理、方法、应用及效果。 large language model ✅
7 Evaluating the Feasibility and Accuracy of Large Language Models for Medical History-Taking in Obstetrics and Gynecology 评估大型语言模型在妇产科病史采集中的可行性和准确性 large language model
8 Integrating Large Language Models with Human Expertise for Disease Detection in Electronic Health Records 利用大型语言模型与人类专业知识相结合，提升电子病历中疾病检测的准确性 large language model
9 Do Large Language Models Exhibit Spontaneous Rational Deception? 大型语言模型在特定情境下会自发进行理性欺骗 large language model
10 Text Chunking for Document Classification for Urban System Management using Large Language Models 提出基于文本分块的LLM方案，用于城市系统管理文档分类，性能媲美人工。 large language model
11 Enhancing Large Language Models (LLMs) for Telecommunications using Knowledge Graphs and Retrieval-Augmented Generation 提出基于知识图谱和检索增强生成（KG-RAG）的框架，提升大型语言模型在电信领域的性能。 large language model
12 BeMERC: Behavior-Aware MLLM-based Framework for Multimodal Emotion Recognition in Conversation BeMERC：基于行为感知的MLLM对话多模态情感识别框架 multimodal
13 Building Instruction-Tuning Datasets from Human-Written Instructions with Open-Weight Large Language Models 利用开源大语言模型和人工指令构建高质量指令微调数据集 large language model
14 Mapping Geopolitical Bias in 11 Large Language Models: A Bilingual, Dual-Framing Analysis of U.S.-China Tensions 双语双框架分析揭示11个大型语言模型在美国-中国议题上的地缘政治偏见。 large language model
15 Large Language Models Pass the Turing Test GPT-4.5通过图灵测试，首次证实大型语言模型具备人类级别的对话能力 large language model
16 Texture or Semantics? Vision-Language Models Get Lost in Font Recognition 揭示视觉-语言模型在字体识别中易受纹理干扰，语义理解能力不足 multimodal chain-of-thought
17 Contradiction Detection in RAG Systems: Evaluating LLMs as Context Validators for Improved Information Consistency 提出RAG系统矛盾检测框架，评估LLM作为上下文验证器的信息一致性能力 large language model chain-of-thought
18 TeleAntiFraud-28k: An Audio-Text Slow-Thinking Dataset for Telecom Fraud Detection 提出TeleAntiFraud-28k：一个用于电信诈骗检测的音频-文本慢思考数据集。 large language model multimodal ✅
19 Contextualize-then-Aggregate: Circuits for In-Context Learning in Gemma-2 2B 通过因果干预揭示Gemma-2 2B模型上下文学习的“语境化-聚合”机制 large language model
20 A Systematic Evaluation of LLM Strategies for Mental Health Text Analysis: Fine-tuning vs. Prompt Engineering vs. RAG 系统评估LLM在心理健康文本分析中的策略：微调、提示工程与RAG large language model
21 Artificial Conversations, Real Results: Fostering Language Detection with Synthetic Data 利用合成数据促进语言检测：一种基于LLM的意大利语包容性语言检测方法 large language model
22 SciReplicate-Bench: Benchmarking LLMs in Agent-driven Algorithmic Reproduction from Research Papers 提出SciReplicate-Bench以评估LLMs在算法重现中的表现 large language model ✅
23 Synthesizing Public Opinions with LLMs: Role Creation, Impacts, and the Future to eDemorcacy 利用大型语言模型合成公众意见，解决传统调查方法中的偏差问题。 large language model
24 Dynamic Parametric Retrieval Augmented Generation for Test-time Knowledge Enhancement 提出动态参数检索增强生成（DyPRAG）框架，解决测试时知识增强的效率与泛化问题。 large language model ✅
25 Adaptive Layer-skipping in Pre-trained LLMs FlexiDepth：一种预训练LLM的自适应层跳跃方法，在Llama-3-8B上实现显著加速。 large language model
26 LANID: LLM-assisted New Intent Discovery LANID：利用LLM辅助轻量级编码器进行新意图发现 large language model ✅
27 Entropy-Based Adaptive Weighting for Self-Training 提出基于熵的自训练自适应加权方法EAST，提升大语言模型数学问题求解能力。 large language model
28 Did ChatGPT or Copilot use alter the style of internet news headlines? A time series regression analysis 研究表明，ChatGPT和Copilot的发布对互联网新闻标题风格的影响有限。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
29 Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learning 提出Rec-R1，通过强化学习桥接生成式大语言模型与用户中心推荐系统。 reinforcement learning distillation large language model
30 TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers' Guidance TwT：通过习惯性推理蒸馏和多教师指导，实现无Token的LLM推理加速 distillation large language model
31 JudgeLRM: Large Reasoning Models as a Judge 提出JudgeLRM，利用强化学习训练面向判断任务的大型推理模型，显著提升推理能力。 reinforcement learning large language model
32 Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains 提出基于可验证奖励的强化学习框架，扩展LLM在多领域自由文本任务中的应用。 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🔗⭐
33 Optimizing Humor Generation in Large Language Models: Temperature Configurations and Architectural Trade-offs 通过温度配置优化大语言模型幽默生成，并分析架构权衡 HuMoR large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
34 Benchmarking Large Language Models for Calculus Problem-Solving: A Comparative Analysis 对比分析大型语言模型在微积分问题求解中的性能表现 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页