cs.CL（2025-06-14）

📊 共 24 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) (20 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 RealFactBench: A Benchmark for Evaluating Large Language Models in Real-World Fact-Checking 提出RealFactBench，用于评估大语言模型在真实世界的事实核查能力 large language model multimodal ✅
2 Enabling Precise Topic Alignment in Large Language Models Via Sparse Autoencoders 提出基于稀疏自编码器的LLM主题对齐方法，无需微调即可实现精确控制。 large language model
3 Intersectional Bias in Japanese Large Language Models from a Contextualized Perspective 提出inter-JBBQ基准，揭示日语大语言模型中基于上下文的交叉性偏见 large language model
4 Exploring Cultural Variations in Moral Judgments with Large Language Models 探讨大语言模型在道德判断中的文化差异 large language model
5 Investigating the Effects of Cognitive Biases in Prompts on Large Language Model Outputs 研究认知偏差对大语言模型输出的影响，揭示提示词偏见与模型可靠性的关系 large language model
6 GSDNet: Revisiting Incomplete Multimodal-Diffusion from Graph Spectrum Perspective for Conversation Emotion Recognition GSDNet：基于图谱视角的对话情感识别不完整多模态扩散模型 multimodal
7 FlexRAG: A Flexible and Comprehensive Framework for Retrieval-Augmented Generation FlexRAG：一个灵活全面的检索增强生成框架，旨在解决现有框架的局限性。 large language model multimodal ✅
8 Profiling News Media for Factuality and Bias Using LLMs and the Fact-Checking Methodology of Human Experts 提出一种基于LLM和事实核查方法的新闻媒体真实性和偏见评估框架 large language model ✅
9 Group then Scale: Dynamic Mixture-of-Experts Multilingual Language Model 提出动态混合专家多语言模型，解决多语言LLM的负迁移问题。 large language model
10 Recent Advances and Future Directions in Literature-Based Discovery 综述文献发掘（LBD）最新进展，聚焦知识图谱、深度学习与大语言模型融合 large language model
11 Refract ICL: Rethinking Example Selection in the Era of Million-Token Models Refract ICL：百万Token模型时代下，重新思考ICL的示例选择策略 large language model
12 Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics 提出基于多维度人物角色的AI辩论框架，研究道德决策与说服动态 large language model
13 Between Predictability and Randomness: Seeking Artistic Inspiration from AI Generative Models 利用AI生成模型探索艺术灵感：LSTM-VAE诗句激发创造力 large language model
14 OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and Metrics OpenUnlearning：通过统一的方法和指标基准测试加速LLM的不可学习性研究 large language model
15 Towards Building General Purpose Embedding Models for Industry 4.0 Agents 构建通用嵌入模型，提升工业4.0智能体在资产维护决策中的语言理解能力。 large language model
16 OneEval: Benchmarking LLM Knowledge-intensive Reasoning over Diverse Knowledge Bases 提出OneEval以解决LLM在知识密集推理中的评估问题 large language model
17 Improving Factuality for Dialogue Response Generation via Graph-Based Knowledge Augmentation 提出基于图知识增强的对话生成框架，提升生成回复的事实性。 large language model
18 TagRouter: Learning Route to LLMs through Tags for Open-Domain Text Generation Tasks 提出TagRouter，通过标签路由LLM，解决开放域文本生成任务中的模型选择问题。 large language model
19 From Outcomes to Processes: Guiding PRM Learning from ORM for Inference-Time Alignment 提出SP-PRM框架，通过过程奖励模型提升LLM推理时对齐效果 large language model
20 Understanding the Effect of Knowledge Graph Extraction Error on Downstream Graph Analyses: A Case Study on Affiliation Graphs 评估知识图谱抽取误差对下游图分析的影响，以机构隶属图为例。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
21 Detection, Classification, and Mitigation of Gender Bias in Large Language Models 提出融合强化学习与思维链的框架，用于检测、分类和缓解大语言模型中的性别偏见。 reinforcement learning DPO direct preference optimization
22 Advances in LLMs with Focus on Reasoning, Adaptability, Efficiency and Ethics 综述大型语言模型在推理、适应性、效率和伦理方面的进展 reinforcement learning large language model multimodal
23 Med-U1: Incentivizing Unified Medical Reasoning in LLMs via Large-scale Reinforcement Learning Med-U1：通过大规模强化学习，激励LLM实现统一的医学推理 reinforcement learning reward design large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
24 Language Surgery in Multilingual Large Language Models 提出Inference-Time Language Control方法，解决多语言LLM中的语言混淆问题。 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页