cs.CL（2024-12-02）

📊 共 23 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) (18 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 MiningGPT -- A Domain-Specific Large Language Model for the Mining Industry MiningGPT：面向矿业的领域特定大语言模型，知识测试得分提升14%。 large language model instruction following
2 The use of large language models to enhance cancer clinical trial educational materials 利用大型语言模型增强癌症临床试验宣教材料的可读性和理解性 large language model
3 Towards Resource Efficient and Interpretable Bias Mitigation in Large Language Models 提出一种资源高效且可解释的偏差缓解方法，通过专家模型在解码时干预LLM输出。 large language model
4 Query Performance Explanation through Large Language Model for HTAP Systems 提出基于LLM的HTAP系统查询性能解释框架，解决跨引擎性能差异理解难题 large language model
5 Adapting Large Language Models to Log Analysis with Interpretable Domain Knowledge 提出SuperLog，通过融入可解释领域知识持续预训练，提升LLM在日志分析任务中的性能。 large language model
6 The "LLM World of Words" English free association norms generated by large language models 构建LLM词语联想规范数据集LWOW，用于研究LLM的知识表征和偏见。 large language model
7 Data Uncertainty-Aware Learning for Multimodal Aspect-based Sentiment Analysis 提出数据不确定性感知学习方法UA-MABSA，提升多模态情感分析在低质量数据上的性能。 multimodal
8 Enhancing Function-Calling Capabilities in LLMs: Strategies for Prompt Formats, Data Integration, and Multilingual Translation 通过优化Prompt、数据集成和多语言翻译增强LLM的函数调用能力 large language model instruction following chain-of-thought
9 NYT-Connections: A Deceptively Simple Text Classification Task that Stumps System-1 Thinkers 提出NYT-Connections基准测试，用于评估LLM的深思熟虑推理能力 large language model chain-of-thought
10 Can We Afford The Perfect Prompt? Balancing Cost and Accuracy with the Economical Prompting Index 提出经济提示指数EPI，在保证精度的前提下，优化大语言模型提示工程的成本。 chain-of-thought
11 If Eleanor Rigby Had Met ChatGPT: A Study on Loneliness in a Post-LLM World 研究表明，通用LLM在非任务导向的孤独场景中存在伦理风险和内容毒性问题 large language model
12 Scaling Law for Language Models Training Considering Batch Size 研究批量大小对大语言模型训练的影响，提出考虑批量大小的缩放定律。 large language model
13 Early Exit Is a Natural Capability in Transformer-based Models: An Empirical Study on Early Exit without Joint Optimization 探索Transformer模型的固有早退能力，无需联合优化即可加速LLM推理。 large language model
14 Berezinskii--Kosterlitz--Thouless transition in a context-sensitive random language model 构建上下文相关的随机语言模型，揭示自然语言中的BKT相变现象 large language model
15 SailCompass: Towards Reproducible and Robust Evaluation for Southeast Asian Languages SailCompass：面向东南亚语言的大语言模型可复现、鲁棒评测基准 large language model
16 Automated Extraction of Acronym-Expansion Pairs from Scientific Papers 提出一种结合正则表达式和大型语言模型的自动化学术论文缩略语-全称对提取方法。 large language model
17 Advancing Speech Language Models by Scaling Supervised Fine-Tuning with Over 60,000 Hours of Synthetic Speech Dialogue Data 通过大规模监督微调，利用6万小时合成语音对话数据，提升语音语言模型性能。 large language model ✅
18 SAUP: Situation Awareness Uncertainty Propagation on LLM Agent 提出SAUP框架，用于LLM Agent多步推理中情境感知的不确定性传播 large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
19 Are We There Yet? Revealing the Risks of Utilizing Large Language Models in Scholarly Peer Review 揭示大语言模型在学术同行评审中应用的风险，强调当前技术尚未成熟 manipulation large language model
20 The Vulnerability of Language Model Benchmarks: Do They Accurately Reflect True LLM Performance? 揭示大语言模型评测基准的脆弱性，质疑其性能反映的真实性 manipulation 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🔗⭐
21 Yi-Lightning Technical Report Yi-Lightning：基于增强型MoE架构的大语言模型，在中文、数学、编程等领域表现卓越。 reinforcement learning RLHF large language model
22 Exploring ReAct Prompting for Task-Oriented Dialogue: Insights and Shortcomings 探索ReAct提示在面向任务对话中的应用及局限性 reinforcement learning large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
23 A Top-down Graph-based Tool for Modeling Classical Semantic Maps: A Crosslinguistic Case Study of Supplementary Adverbs 提出一种基于图的自顶向下方法，用于自动构建跨语言语义地图。 semantic map ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页