cs.CL（2024-07-15）

📊 共 28 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱一：机器人控制 (Robot Control) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 Empowering Persian LLMs for Instruction Following: A Novel Dataset and Training Approach 提出FarsInstruct波斯语指令数据集和Co-CoLA训练框架，提升波斯语LLM的指令遵循能力。 large language model instruction following
2 Bridging Sequence-Structure Alignment in RNA Foundation Models OmniGenome：提出一种RNA基础模型，通过结构上下文建模对齐序列-结构，实现RNA序列和结构的双向映射。 foundation model
3 Q-Sparse: All Large Language Models can be Fully Sparsely-Activated Q-Sparse：实现大语言模型全稀疏激活，提升推理效率 large language model
4 MMM: Multilingual Mutual Reinforcement Effect Mix Datasets & Test with Open-domain Information Extraction Large Language Models 提出多语言互增强效应混合数据集MMM，并用于训练开放域信息抽取大语言模型OIELLM。 large language model ✅
5 An Actionable Framework for Assessing Bias and Fairness in Large Language Model Use Cases 提出LLM偏见评估框架LangFair，针对特定用例评估模型公平性 large language model
6 Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation 提出FLAMe：通过训练基础大语言模型提升自动评估能力 large language model
7 Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation 提出Think-on-Graph 2.0，通过知识图谱引导的检索增强生成实现深度LLM推理。 large language model ✅
8 Graphusion: Leveraging Large Language Models for Scientific Knowledge Graph Fusion and Construction in NLP Education Graphusion：利用大语言模型融合科学知识图谱，应用于NLP教育 large language model
9 Evaluating Large Language Models with fmeval fmeval：一个用于评估大型语言模型性能和负责任AI维度的开源库 large language model ✅
10 Prompt Selection Matters: Enhancing Text Annotations for Social Sciences with Large Language Models 针对社会科学文本标注，提出基于自动Prompt优化的LLM标注方法，显著提升标注精度。 large language model
11 MetaTool: Facilitating Large Language Models to Master Tools with Meta-task Augmentation MetaTool：通过元任务增强提升大语言模型工具使用能力 large language model
12 TCM-FTP: Fine-Tuning Large Language Models for Herbal Prescription Prediction TCM-FTP：通过微调大型语言模型进行中药处方预测 large language model
13 Multilingual Contrastive Decoding via Language-Agnostic Layers Skipping 提出基于语言无关层跳跃的多语言对比解码方法，提升LLM在多语言推理任务中的性能。 large language model chain-of-thought ✅
14 Qwen2 Technical Report Qwen2系列发布：开源0.5B-72B参数规模语言模型，性能超越现有开源模型 large language model multimodal
15 GraphEval: A Knowledge-Graph Based LLM Hallucination Evaluation Framework GraphEval：一种基于知识图谱的LLM幻觉评估框架 large language model
16 Codebook LLMs: Evaluating LLMs as Measurement Tools for Political Science Concepts 提出Codebook-LLM框架，评估LLM在政治科学概念测量中的应用，并提供改进指导。 large language model
17 CLAVE: An Adaptive Framework for Evaluating Values of LLM Generated Responses CLAVE：一种自适应框架，用于评估LLM生成响应的价值观 large language model
18 DOCBENCH: A Benchmark for Evaluating LLM-based Document Reading Systems 提出DocBench：用于评估基于LLM的文档阅读系统的基准 large language model
19 Enhancing Retrieval and Managing Retrieval: A Four-Module Synergy for Improved Quality and Efficiency in RAG Systems 提出ERM4框架，通过四个模块协同优化RAG系统的质量和效率。 large language model ✅
20 An Empirical Study of Validating Synthetic Data for Formula Generation 通过验证合成数据提升公式生成模型性能 large language model
21 Beyond Generative Artificial Intelligence: Roadmap for Natural Language Generation 自然语言生成发展路线图：应对大型语言模型时代的新挑战 large language model
22 How and where does CLIP process negation? 深入剖析CLIP如何处理否定概念，揭示多模态模型内部机制 multimodal
23 The Good, The Bad, and The Greedy: Evaluation of LLMs Should Not Ignore Non-Determinism 关注LLM非确定性：揭示贪婪解码与采样策略的性能差异及影响 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🔗⭐
24 Leave No Knowledge Behind During Knowledge Distillation: Towards Practical and Effective Knowledge Distillation for Code-Switching ASR Using Realistic Data 提出K²D方法，利用无标注数据蒸馏，提升代码切换语音识别效率与精度。 distillation foundation model ✅
25 Mix-CPT: A Domain Adaptation Framework via Decoupling Knowledge Learning and Format Alignment 提出Mix-CPT框架以解决LLM领域适应问题 distillation large language model
26 Qwen2-Audio Technical Report Qwen2-Audio：基于自然语言提示的大规模音频语言模型，实现语音交互与音频分析 DPO instruction following
27 Don't Throw Away Data: Better Sequence Knowledge Distillation 提出基于MBR的多样性序列知识蒸馏方法，提升机器翻译性能。 distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
28 Uncertainty is Fragile: Manipulating Uncertainty in Large Language Models 提出针对大语言模型不确定性估计的后门攻击，可操纵模型置信度。 manipulation large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页