cs.CL（2026-02-23）

📊 共 16 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Personalized Prediction of Perceived Message Effectiveness Using Large Language Model Based Digital Twins	利用大语言模型数字孪生进行个性化消息有效性预测，提升移动健康干预效果	large language model
2	To Reason or Not to: Selective Chain-of-Thought in Medical Question Answering	提出选择性思维链（Selective CoT）方法，提升医学问答效率并降低计算成本。	large language model chain-of-thought
3	Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval	首个多模态文档智能综述：聚焦视觉文档检索与多模态大语言模型	large language model multimodal
4	Multilingual Large Language Models do not comprehend all natural languages to equal degrees	揭示多语言大模型对不同自然语言理解能力差异，挑战英语最佳表现的预设	large language model
5	Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming	提出基于模拟的临床红队测试框架，评估大语言模型在心理健康支持中的风险	large language model
6	Entropy in Large Language Models	通过熵分析比较大型语言模型与自然语言的差异	large language model
7	Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework	提出Prune-then-Merge框架以解决多向量视觉文档检索效率问题	multimodal
8	NanoKnow: How to Know What Your Language Model Knows	NanoKnow：构建基准数据集，探究LLM参数知识来源及外部知识互补性	large language model	✅
9	Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference	提出Pyramid MoA，通过动态路由降低大语言模型推理成本，提升性价比。	large language model
10	ReAttn: Improving Attention-based Re-ranking via Attention Re-weighting	提出ReAttn：通过注意力重加权改进基于注意力的重排序方法	large language model
11	Position: General Alignment Has Hit a Ceiling; Edge Alignment Must Be Taken Seriously	提出边缘对齐，解决通用对齐在复杂社会技术系统中存在的局限性	large language model
12	gencat: Generative computerized adaptive testing	提出GENCAT：一种利用生成式大语言模型的自适应测试框架	large language model
13	SAMAS: A Spectrum-Guided Multi-Agent System for Achieving Style Fidelity in Literary Translation	提出SAMAS，通过频谱引导的多Agent系统提升文学翻译中的风格保真度。	large language model
14	KGHaluBench: A Knowledge Graph-Based Hallucination Benchmark for Evaluating the Breadth and Depth of LLM Knowledge	KGHaluBench：基于知识图谱的大语言模型幻觉评测基准，评估知识的广度和深度	large language model
15	Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning	提出DUAL基准以解决机器遗忘中的知识来源问题	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
16	How to Train Your Deep Research Agent? Prompt, Reward, and Policy Optimization in Search-R1	针对深度研究Agent，系统性研究Prompt、奖励函数和策略优化方法，提升Search-R1性能。	reinforcement learning PPO

⬅️ 返回 cs.CL 首页 · 🏠 返回主页