cs.CL（2024-07-02）

📊 共 31 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) (27 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (4)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (27 篇)

#题目一句话要点标签🔗⭐
1 Towards a Holistic Framework for Multimodal Large Language Models in Three-dimensional Brain CT Report Generation 提出BrainGPT，一个用于生成3D脑部CT报告的临床视觉指令调优多模态大语言模型框架 large language model multimodal
2 Open foundation models for Azerbaijani language 为阿塞拜疆语构建开源基础模型，并进行系统性评测 large language model foundation model
3 Synthetic Multimodal Question Generation 提出SMMQG框架，用于合成多模态问答数据，解决MMRAG评估数据匮乏问题。 large language model multimodal
4 LLMs Plagiarize: Ensuring Responsible Sourcing of Large Language Model Training Data Through Knowledge Graph Comparison 提出基于知识图谱比较的剽窃检测系统，用于评估LLM训练数据来源的合规性。 large language model
5 CEB: Compositional Evaluation Benchmark for Fairness in Large Language Models 提出CEB：用于评估大型语言模型公平性的组合评估基准 large language model
6 Is Your AI-Generated Code Really Safe? Evaluating Large Language Models on Secure Code Generation with CodeSecEval CodeSecEval：评估并提升大语言模型在安全代码生成方面的能力 large language model
7 Generative Large Language Models in Automated Fact-Checking: A Survey 综述生成式大语言模型在自动化事实核查中的应用 large language model
8 CFinBench: A Comprehensive Chinese Financial Benchmark for Large Language Models CFinBench：一个全面的中文金融领域大语言模型评测基准 large language model ✅
9 Predictive Simultaneous Interpretation: Harnessing Large Language Models for Democratizing Real-Time Multilingual Communication 利用大型语言模型预测能力，实现预测性同声传译 large language model
10 PromptIntern: Saving Inference Costs by Internalizing Recurrent Prompt during Large Language Model Fine-tuning PromptIntern：通过内化循环Prompt来节省大语言模型微调的推理成本 large language model
11 Generative Monoculture in Large Language Models 揭示大语言模型中的生成单一种植现象，并探讨其潜在风险与缓解策略 large language model
12 WTU-EVAL: A Whether-or-Not Tool Usage Evaluation Benchmark for Large Language Models 提出WTU-Eval基准，评估大语言模型在不确定场景下的工具使用能力 large language model
13 Automatic Adaptation Rule Optimization via Large Language Models 利用大型语言模型优化自适应规则，提升系统性能与鲁棒性 large language model
14 Crossroads of Continents: Automated Artifact Extraction for Cultural Adaptation with Large Multimodal Models 提出CultureAdapt，利用大型多模态模型自动提取文化制品并实现文化适应。 multimodal ✅
15 MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention MInference：通过动态稀疏注意力加速长文本LLM的预填充阶段 large language model
16 Learning to Refine with Fine-Grained Natural Language Feedback 提出DCR框架，利用细粒度自然语言反馈提升LLM生成的事实一致性。 large language model
17 Pelican: Correcting Hallucination in Vision-LLMs via Claim Decomposition and Program of Thought Verification Pelican：通过声明分解和思维程序验证来纠正视觉-LLM中的幻觉问题 instruction following
18 MORPHEUS: Modeling Role from Personalized Dialogue History by Exploring and Utilizing Latent Space 提出MORPHEUS框架，通过探索和利用潜在空间建模个性化对话历史中的角色信息。 large language model
19 Why do LLaVA Vision-Language Models Reply to Images in English? LLaVA模型在多模态问答中存在显著的英语偏见问题，本研究揭示其内在原因并提出改进方向。 multimodal
20 Efficient Sparse Attention needs Adaptive Token Release 提出自适应Token释放的稀疏注意力机制，提升LLM推理效率。 large language model ✅
21 Exploring the Role of Transliteration in In-Context Learning for Low-resource Languages Written in Non-Latin Scripts 探索音译在非拉丁文字低资源语言LLM上下文学习中的作用 large language model
22 Evaluating the Ability of LLMs to Solve Semantics-Aware Process Mining Tasks 评估LLM在语义感知的过程挖掘任务中的能力，并提出微调方案。 large language model
23 Assessing the Effectiveness of GPT-4o in Climate Change Evidence Synthesis and Systematic Assessments: Preliminary Insights 评估GPT-4o在气候变化证据综合和系统评估中的有效性 large language model
24 Whispering Experts: Neural Interventions for Toxicity Mitigation in Language Models 提出AurA干预方法，通过神经元毒性判别能力降低大语言模型的有害内容生成 large language model
25 GemmAr: Enhancing LLMs Through Arabic Instruction-Tuning GemmAr：通过阿拉伯语指令微调增强大型语言模型 large language model
26 Breaking Language Barriers: Cross-Lingual Continual Pre-Training at Scale 提出跨语言持续预训练方法，突破大语言模型的多语言扩展壁垒 large language model
27 Helpful assistant or fruitful facilitator? Investigating how personas affect language model behavior 研究角色对大型语言模型行为的影响 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🔗⭐
28 Towards Human Understanding of Paraphrase Types in Large Language Models 提出APTY数据集，评估大型语言模型在原子释义类型上的理解能力 RLHF DPO large language model
29 RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs 提出一种可扩展的多语言偏好优化方法，显著提升LLM在多种语言上的对齐效果。 RLHF large language model
30 Talking to Machines: do you read me? 综述式对话系统研究：从模块化架构到端到端深度神经网络 reinforcement learning large language model multimodal
31 Predicting vs. Acting: A Trade-off Between World Modeling & Agent Modeling RLHF模型在生成任务中存在预测能力与行为能力之间的权衡 RLHF world model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页