cs.CL（2025-09-21）

📊 共 12 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) (9 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Probabilistic Token Alignment for Large Language Model Fusion	提出基于概率Token对齐的PTA-LLM，用于解决大语言模型融合中的词汇对齐问题	large language model
2	LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization	LifeAlign：面向大语言模型的终身对齐与记忆增强的聚焦偏好优化	large language model
3	Uncovering Implicit Bias in Large Language Models with Concept Learning Dataset	提出概念学习数据集，揭示大语言模型中量词单调性的隐式偏见	large language model
4	Context Is What You Need: The Maximum Effective Context Window for Real World Limits of LLMs	揭示大语言模型上下文窗口的真实有效性：有效上下文窗口远小于理论上限	large language model
5	TactfulToM: Do LLMs Have the Theory of Mind Ability to Understand White Lies?	提出TactfulToM基准，评估LLM在理解善意谎言中的心理理论能力	large language model
6	Influence Guided Context Selection for Effective Retrieval-Augmented Generation	提出基于上下文影响值引导的上下文选择方法，提升检索增强生成效果。	large language model	✅
7	Attention Consistency for LLMs Explanation	提出MACS，通过注意力一致性提升LLM解释性并降低计算成本	large language model
8	Modeling Bottom-up Information Quality during Language Processing	提出基于互信息的阅读理解模型，研究视觉信息质量对语言处理的影响	multimodal
9	K-DeCore: Facilitating Knowledge Transfer in Continual Structured Knowledge Reasoning via Knowledge Decoupling	K-DeCore：通过知识解耦促进持续结构化知识推理中的知识迁移	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	🔗	⭐
10	Preference Distillation via Value based Reinforcement Learning	提出基于价值强化学习的偏好蒸馏方法TVKD，提升小模型DPO训练效果。	reinforcement learning behavior cloning DPO
11	Can GRPO Boost Complex Multimodal Table Understanding?	Table-R1：通过三阶段强化学习提升复杂多模态表格理解能力	reinforcement learning multimodal
12	CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification	提出HiDAC模型，用于解决跨框架多语篇章关系分类任务的挑战。	contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页