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) (14) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 English Pronunciation Evaluation without Complex Joint Training: LoRA Fine-tuned Speech Multimodal LLM 提出基于LoRA微调的多模态LLM，用于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 From Construction to Injection: Edit-Based Fingerprints for Large Language Models 提出基于编辑的指纹框架，用于保护大型语言模型的版权和防止未经授权的传播。 large language model
4 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
5 Beyond ROUGE: N-Gram Subspace Features for LLM Hallucination Detection 提出基于N-Gram子空间的LLM幻觉检测方法，显著提升检测精度。 large language model
6 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
7 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。 large language model
8 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，构建分析数据集 large language model
9 SESGO: Spanish Evaluation of Stereotypical Generative Outputs SESGO：提出一种文化敏感的西班牙语刻板印象生成评估框架，揭示现有LLM的偏见。 large language model
10 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
11 Measuring Scalar Constructs in Social Science with LLMs 提出基于LLM的社会科学标量构建测量方法，提升文本量化分析的准确性。 large language model
12 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
13 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，揭示语言模型跨语言性能差异 large language model
14 Mitigation of Gender and Ethnicity Bias in AI-Generated Stories through Model Explanations 提出BAME方法，通过模型解释缓解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🔗⭐
15 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，显著提升游戏完成率和胜率。 reinforcement learning deep reinforcement learning world model
16 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
17 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
18 Training LLMs to be Better Text Embedders through Bidirectional Reconstruction 提出双向重构训练方法，提升LLM文本嵌入在检索和重排序任务中的性能 contrastive learning 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以解决多文献科学推理问题 PULSE ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页