cs.CL（2025-10-24）

📊 共 18 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) (15 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱一：机器人控制 (Robot Control) (1)

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

#	题目	一句话要点	标签	🔗	⭐
1	Emotions Where Art Thou: Understanding and Characterizing the Emotional Latent Space of Large Language Models	探索大型语言模型的情感表征：揭示并操控其潜在情感空间	large language model
2	Toward Understanding the Transferability of Adversarial Suffixes in Large Language Models	分析对抗后缀在大型语言模型中的可迁移性，并提出提升攻击成功率的方法	large language model
3	A Stylometric Application of Large Language Models	利用大型语言模型进行文体分析，区分不同作者的写作风格。	large language model
4	Doc-Researcher: A Unified System for Multimodal Document Parsing and Deep Research	提出Doc-Researcher，用于多模态文档解析和深度研究的统一系统。	multimodal
5	ColorEcosystem: Powering Personalized, Standardized, and Trustworthy Agentic Service in massive-agent Ecosystem	ColorEcosystem：赋能大规模Agent生态系统中个性化、标准化和可信赖的Agent服务	large language model multimodal	✅
6	Optimal Detection for Language Watermarks with Pseudorandom Collision	针对语言水印检测中伪随机碰撞问题，提出最优检测框架	large language model
7	AI-Mediated Communication Reshapes Social Structure in Opinion-Diverse Groups	AI辅助沟通重塑意见群体中的社会结构，影响群体凝聚与分裂	large language model
8	Uncovering the Persuasive Fingerprint of LLMs in Jailbreaking Attacks	利用社会科学说服理论，提升LLM越狱攻击的成功率	large language model
9	Model-Aware Tokenizer Transfer	提出模型感知的分词器迁移方法MATT，提升低资源语言LLM性能。	large language model
10	Confidence is Not Competence	揭示大语言模型置信度与能力脱钩机制：几何复杂度差异解释	large language model
11	The Universal Landscape of Human Reasoning	提出IF-Track，利用大语言模型量化人类推理过程中的信息流动，统一建模推理行为。	large language model
12	From Polyester Girlfriends to Blind Mice: Creating the First Pragmatics Understanding Benchmarks for Slovene	为斯洛文尼亚语提出首个语用理解基准测试集SloPragEval和SloPragMega	large language model
13	Are the LLMs Capable of Maintaining at Least the Language Genus?	研究LLM是否能维持语言谱系关系，揭示训练数据对多语言能力的影响	large language model
14	Deep Literature Survey Automation with an Iterative Workflow	提出IterSurvey，通过迭代式大纲生成实现高质量的文献综述自动化。	multimodal	✅
15	Wisdom and Delusion of LLM Ensembles for Code Generation and Repair	提出基于多样性的LLM集成方法，显著提升代码生成与修复性能。	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
16	Compositional Bias Control in Large Language Models: Preference Learning Fails, Supervision Succeeds	对比研究表明，监督微调优于偏好学习，能有效控制大语言模型中的组合偏见。	preference learning DPO direct preference optimization
17	RETuning: Upgrading Inference-Time Scaling for Stock Movement Prediction with Large Language Models	提出RETuning方法，提升大语言模型在股票预测中的推理能力	reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
18	Document Understanding, Measurement, and Manipulation Using Category Theory	利用范畴论进行文档理解、测量和操作，实现文档总结与扩展。	manipulation multimodal

⬅️ 返回 cs.CL 首页 · 🏠 返回主页