cs.CL（2025-09-03）

📊 共 19 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) (13) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱一：机器人控制 (Robot Control) (1) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 English Pronunciation Evaluation without Complex Joint Training: LoRA Fine-tuned Speech Multimodal LLM 利用LoRA微调多模态LLM实现高效英语发音评估与诊断 large language model multimodal
2 ProMQA-Assembly: Multimodal Procedural QA Dataset on Assembly 提出ProMQA-Assembly多模态程序化问答数据集，用于评估装配任务助手。 multimodal
3 Structure-Learnable Adapter Fine-Tuning for Parameter-Efficient Large Language Models 提出结构可学习的Adapter微调方法，提升大语言模型参数效率和任务适应性 large language model
4 Beyond ROUGE: N-Gram Subspace Features for LLM Hallucination Detection 提出基于N-Gram子空间特征的LLM幻觉检测方法，显著提升检测性能。 large language model
5 NoteBar: An AI-Assisted Note-Taking System for Personal Knowledge Management NoteBar：一种AI辅助的笔记系统，用于个人知识管理 large language model
6 Curse of Knowledge: When Complex Evaluation Context Benefits yet Biases LLM Judges 构建ComplexEval基准，揭示并量化LLM评判在复杂评估中存在的辅助信息诱导偏差问题。 large language model
7 Using LLMs to create analytical datasets: A case study of reconstructing the historical memory of Colombia 利用大型语言模型重建哥伦比亚历史记忆，创建分析数据集。 large language model
8 SESGO: Spanish Evaluation of Stereotypical Generative Outputs SESGO：提出西班牙语刻板印象生成输出评估框架，填补多语言LLM偏见评估的空白。 large language model
9 Domain Adaptation of LLMs for Process Data 提出基于LLM领域自适应的过程数据预测方法，提升预测过程监控性能 large language model
10 Measuring Scalar Constructs in Social Science with LLMs 利用LLM测量社会科学中的标量结构，提出token概率加权评分方法并验证其有效性。 large language model
11 DiaCBT: A Long-Periodic Dialogue Corpus Guided by Cognitive Conceptualization Diagram for CBT-based Psychological Counseling DiaCBT：构建认知概念化图引导的CBT心理咨询长周期对话语料库 large language model
12 Artificially Fluent: Swahili AI Performance Benchmarks Between English-Trained and Natively-Trained Datasets 对比英语训练与斯瓦希里语原生训练，揭示LLM跨语言性能差异 large language model
13 Mitigation of Gender and Ethnicity Bias in AI-Generated Stories through Model Explanations 提出BAME方法，利用模型解释缓解AI生成故事中的性别和种族偏见。 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🔗⭐
14 Design and Optimization of Reinforcement Learning-Based Agents in Text-Based Games 提出一种基于强化学习的文本游戏智能体设计与优化方法 reinforcement learning deep reinforcement learning world model
15 Advancing SLM Tool-Use Capability using Reinforcement Learning 利用强化学习GRPO提升小语言模型（SLM）的工具使用能力 reinforcement learning large language model
16 Breaking the Mirror: Activation-Based Mitigation of Self-Preference in LLM Evaluators 提出基于激活的干预方法，缓解LLM评估器中的自我偏好问题 direct preference optimization large language model
17 Training LLMs to be Better Text Embedders through Bidirectional Reconstruction 提出双向重建训练方法，提升LLM作为文本嵌入模型的性能 contrastive 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 From Injection to Defense: Constructing Edit-Based Fingerprints for Large Language Models 提出RFEdit框架，通过知识编辑为大语言模型构建基于编辑的指纹，并提出FSFT进行防御。 manipulation large language model

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
19 ResearchPulse: Building Method-Experiment Chains through Multi-Document Scientific Inference 提出ResearchPulse，通过多文档科学推理构建方法-实验链，促进科研发展脉络理解。 PULSE ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页