cs.CL（2024-05-25）

📊 共 16 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗1) 支柱一：机器人控制 (Robot Control) (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 Generating clickbait spoilers with an ensemble of large language models 提出一种基于大型语言模型集成的方法，用于生成反点击诱饵剧透文本。 large language model
2 Confidence Under the Hood: An Investigation into the Confidence-Probability Alignment in Large Language Models 提出信心-概率对齐概念以提升大语言模型的可靠性 large language model
3 ConStat: Performance-Based Contamination Detection in Large Language Models ConStat：基于性能的大语言模型污染检测方法，有效识别并量化模型作弊行为 large language model
4 5W1H Extraction With Large Language Models 提出基于LLM的5W1H要素抽取方法，并构建高质量数据集以提升模型性能。 large language model
5 Towards Completeness-Oriented Tool Retrieval for Large Language Models 提出COLT，一种面向完备性的工具检索方法，提升大语言模型工具增强效果 large language model
6 SPP: Sparsity-Preserved Parameter-Efficient Fine-Tuning for Large Language Models SPP：一种稀疏保持的参数高效微调方法，用于提升稀疏大语言模型的性能。 large language model ✅
7 Incremental Comprehension of Garden-Path Sentences by Large Language Models: Semantic Interpretation, Syntactic Re-Analysis, and Attention 利用大型语言模型增量理解花园路径句：语义解释、句法重分析与注意力机制 large language model
8 Picturing Ambiguity: A Visual Twist on the Winograd Schema Challenge WinoVis：一个用于评估文本到图像模型在指代消解能力上的视觉挑战数据集 large language model multimodal
9 STRIDE: A Tool-Assisted LLM Agent Framework for Strategic and Interactive Decision-Making STRIDE：一种工具辅助的LLM Agent框架，用于战略和交互式决策 large language model
10 Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Tracking 提出RA-Rec：一种基于检索增强和提示学习的对话式推荐系统，用于解决用户意图理解难题。 large language model
11 No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks 揭示微调攻击的不同机制：EHA与ISA攻击大型语言模型安全性的差异性分析 large language model
12 Towards Unlocking Insights from Logbooks Using AI 提出基于RAG的AI模型，提升粒子加速器电子日志的可访问性和可用性。 multimodal
13 Accelerating Inference of Retrieval-Augmented Generation via Sparse Context Selection 提出Sparse RAG，通过稀疏上下文选择加速检索增强生成推理。 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🔗⭐
14 Keypoint-based Progressive Chain-of-Thought Distillation for LLMs 提出基于关键点的渐进式思维链蒸馏方法KPOD，提升小模型推理能力。 distillation large language model chain-of-thought
15 Improving Multi-lingual Alignment Through Soft Contrastive Learning 提出基于软对比学习的多语言对齐方法，提升跨语言表征质量 contrastive learning ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
16 Learning to Reason via Program Generation, Emulation, and Search 提出CoGEX，通过程序生成、模拟执行和搜索，扩展语言模型的推理能力。 manipulation ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页