cs.CL（2026-02-22）

📊 共 11 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (10 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models	揭示大语言模型中的价值纠缠现象：不同类型价值表征的混淆	large language model
2	Retrieval Augmented Enhanced Dual Co-Attention Framework for Target Aware Multimodal Bengali Hateful Meme Detection	提出检索增强的增强双重协同注意力框架xDORA，用于识别孟加拉语仇恨模因。	multimodal
3	Astra: Activation-Space Tail-Eigenvector Low-Rank Adaptation of Large Language Models	Astra：利用激活空间尾部特征向量的低秩适配方法，提升大语言模型微调性能	large language model
4	Do LLMs and VLMs Share Neurons for Inference? Evidence and Mechanisms of Cross-Modal Transfer	提出SNRF框架，通过共享神经元低秩融合，提升LVLM多步推理能力。	large language model multimodal	✅
5	Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations	剖析Agentic Memory：构建分类体系并实证分析评估与系统局限性	large language model
6	Next Reply Prediction X Dataset: Linguistic Discrepancies in Naively Generated Content	提出基于X平台的新型回复预测数据集，评估LLM生成内容与人类表达的语言差异。	large language model
7	How Do LLMs Encode Scientific Quality? An Empirical Study Using Monosemantic Features from Sparse Autoencoders	利用稀疏自编码器的单义特征，研究LLM如何编码科学质量	large language model
8	Facet-Level Persona Control by Trait-Activated Routing with Contrastive SAE for Role-Playing LLMs	提出基于对比SAE和特征激活路由的角色扮演LLM性格控制方法	large language model
9	A Dataset for Named Entity Recognition and Relation Extraction from Art-historical Image Descriptions	提出FRAME数据集，用于艺术史图像描述的命名实体识别和关系抽取。	large language model
10	Uncovering Context Reliance in Unstructured Knowledge Editing	提出COIN框架，解决非结构化知识编辑中LLM对上下文的过度依赖问题	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	🔗	⭐
11	IAPO: Information-Aware Policy Optimization for Token-Efficient Reasoning	提出IAPO，通过信息论优化策略，提升大模型推理效率并缩短推理链。	reward shaping large language model	✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页