cs.CL（2025-01-21）

📊 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) (3)

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

#题目一句话要点标签🔗⭐
1 Zero-Shot Verification-guided Chain of Thoughts 提出基于零样本验证引导的思维链方法，提升LLM推理能力 large language model chain-of-thought
2 Episodic Memories Generation and Evaluation Benchmark for Large Language Models 提出LLM情景记忆生成与评估基准，揭示现有模型在复杂时空推理上的不足。 large language model
3 Can open source large language models be used for tumor documentation in Germany? -- An evaluation on urological doctors' notes 评估开源大语言模型在德国泌尿科肿瘤文档自动生成中的应用潜力 large language model ✅
4 Leveraging Graph Structures and Large Language Models for End-to-End Synthetic Task-Oriented Dialogues GraphTOD：利用图结构和大型语言模型实现端到端合成任务型对话生成 large language model
5 A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models 提出GraphRAG，通过图结构知识增强定制化大语言模型，解决专业领域知识集成难题。 large language model ✅
6 A Hybrid Attention Framework for Fake News Detection with Large Language Models 提出一种基于混合注意力机制和大型语言模型的假新闻检测框架 large language model
7 Proverbs Run in Pairs: Evaluating Proverb Translation Capability of Large Language Model 评估大语言模型谚语翻译能力，揭示文化元素翻译的挑战与评估难题 large language model
8 Benchmarking Generative AI for Scoring Medical Student Interviews in Objective Structured Clinical Examinations (OSCEs) 利用生成式AI评估医学学生OSCE面试表现，实现客观结构化临床考试评分自动化。 large language model chain-of-thought
9 Automatic Labelling with Open-source LLMs using Dynamic Label Schema Integration 提出基于动态标签模式集成的检索增强分类(RAC)，提升开源LLM自动标注性能。 large language model
10 AdaServe: Accelerating Multi-SLO LLM Serving with SLO-Customized Speculative Decoding AdaServe：通过服务等级目标定制的推测解码加速多服务等级目标LLM服务 large language model
11 Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities 提出BIDS算法，通过平衡数据选择提升指令微调后LLM的多样能力 large language model
12 Revealing emergent human-like conceptual representations from language prediction 揭示：大型语言模型通过语言预测涌现类人概念表征 large language model
13 Condor: Enhance LLM Alignment with Knowledge-Driven Data Synthesis and Refinement Condor：利用知识驱动的数据合成与优化提升LLM对齐 large language model
14 TAD-Bench: A Comprehensive Benchmark for Embedding-Based Text Anomaly Detection TAD-Bench：一个全面的基于嵌入的文本异常检测基准 large language model
15 From Drafts to Answers: Unlocking LLM Potential via Aggregation Fine-Tuning 提出聚合微调（AFT）方法，通过学习整合草稿答案提升大语言模型性能。 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🔗⭐
16 Preference Curriculum: LLMs Should Always Be Pretrained on Their Preferred Data 提出基于困惑度差异的偏好课程学习框架，提升LLM预训练效果 curriculum learning large language model
17 Panoramic Interests: Stylistic-Content Aware Personalized Headline Generation 提出SCAPE框架，解决个性化新闻标题生成中用户风格偏好被忽略的问题。 contrastive learning large language model
18 Generating Plausible Distractors for Multiple-Choice Questions via Student Choice Prediction 提出基于学生选择预测的干扰项生成方法，提升多选题的区分度。 DPO direct preference optimization

⬅️ 返回 cs.CL 首页 · 🏠 返回主页