cs.CL（2025-05-01）

📊 共 19 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) (4)

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

#题目一句话要点标签🔗⭐
1 Reasoning Capabilities and Invariability of Large Language Models 提出几何图形推理基准，评估大型语言模型的逻辑推理能力和提示依赖性 large language model chain-of-thought
2 Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Models 提出SAGE框架，用于评估大语言模型的高阶社会认知能力 large language model
3 Steering Large Language Models with Register Analysis for Arbitrary Style Transfer 利用Register分析引导大语言模型实现任意风格迁移 large language model
4 Large Language Models Understanding: an Inherent Ambiguity Barrier 论证大型语言模型理解能力存在内在模糊性壁垒 large language model
5 Block Circulant Adapter for Large Language Models 提出基于分块循环矩阵适配器的LLM微调方法，降低存储和计算成本。 large language model
6 A Comparative Study of Large Language Models and Human Personality Traits 研究表明大型语言模型人格特质具有动态性和输入依赖性，并提出分布式人格框架。 large language model
7 Triggering Hallucinations in LLMs: A Quantitative Study of Prompt-Induced Hallucination in Large Language Models 提出一种基于提示的框架，用于触发和量化大型语言模型中的幻觉现象。 large language model
8 Red Teaming Large Language Models for Healthcare 通过红队测试发现大型语言模型在医疗领域的潜在危害 large language model
9 Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics 针对LLM Agent记忆机制，提出包含表示、操作和新兴主题的系统性分类框架。 large language model ✅
10 On the generalization of language models from in-context learning and finetuning: a controlled study 研究表明上下文学习比微调在泛化方面更灵活，并提出加入推理轨迹以提升微调泛化能力。 large language model
11 The Illusion of Role Separation: Hidden Shortcuts in LLM Role Learning (and How to Fix Them) 揭示LLM角色学习中的隐藏捷径，并提出基于不变信号强化的解决方案 large language model
12 FreqKV: Frequency Domain Key-Value Compression for Efficient Context Window Extension FreqKV：提出频域Key-Value压缩方法，高效扩展LLM上下文窗口 large language model
13 MoxE: Mixture of xLSTM Experts with Entropy-Aware Routing for Efficient Language Modeling MoxE：结合xLSTM专家混合模型与熵感知路由，提升语言建模效率 large language model
14 LLM-Based Human-Agent Collaboration and Interaction Systems: A Survey 首个LLM驱动的人机协作与交互系统综述，提升智能体可靠性与安全性。 large language model ✅
15 KoACD: The First Korean Adolescent Dataset for Cognitive Distortion Analysis via Role-Switching Multi-LLM Negotiation 提出KoACD：首个面向韩国青少年认知扭曲分析的大规模数据集，并采用多LLM协商方法提升标注质量。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
16 DeepCritic: Deliberate Critique with Large Language Models DeepCritic：利用大语言模型进行数学问题求解过程的深度批判，提升反馈质量。 reinforcement learning large language model
17 FineScope : Precision Pruning for Domain-Specialized Large Language Models Using SAE-Guided Self-Data Cultivation FineScope：利用SAE引导的自数据生成，实现领域专用大语言模型的精确剪枝 distillation large language model
18 CSE-SFP: Enabling Unsupervised Sentence Representation Learning via a Single Forward Pass 提出CSE-SFP，通过单次前向传播实现生成式PLM的无监督句子表示学习 representation learning contrastive learning
19 100 Days After DeepSeek-R1: A Survey on Replication Studies and More Directions for Reasoning Language Models 总结DeepSeek-R1复现研究，探索推理语言模型SFT与RLVR的优化方向 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页