cs.CL（2024-10-03）

📊 共 33 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 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1)

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

#题目一句话要点标签🔗⭐
1 Large Language Model for Multi-Domain Translation: Benchmarking and Domain CoT Fine-tuning 提出领域链式思维微调以解决多领域翻译问题 large language model chain-of-thought
2 Measuring and Improving Persuasiveness of Large Language Models 提出PersuasionBench和PersuasionArena，用于评估和提升大型语言模型的说服力。 large language model
3 MedVisionLlama: Leveraging Pre-Trained Large Language Model Layers to Enhance Medical Image Segmentation MedVisionLlama：利用预训练LLM层增强医学图像分割 large language model ✅
4 Better Instruction-Following Through Minimum Bayes Risk 通过最小贝叶斯风险提升指令跟随能力，并利用自训练降低推理成本。 instruction following
5 Unified Multimodal Interleaved Document Representation for Retrieval 提出统一多模态交错文档表示方法，用于提升检索任务性能。 multimodal
6 Undesirable Memorization in Large Language Models: A Survey 综述大型语言模型中的不良记忆化现象，分析其风险与应对策略。 large language model
7 Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers 提出In-Context Re-ranking (ICR)，利用LLM注意力机制实现高效零样本重排序。 large language model
8 Defining Knowledge: Bridging Epistemology and Large Language Models 探讨LLM知识定义：桥接认知论与大语言模型，提出评估协议 large language model
9 Domain-Specific Retrieval-Augmented Generation Using Vector Stores, Knowledge Graphs, and Tensor Factorization 提出SMART-SLIC框架，结合RAG、知识图谱和向量存储，提升领域特定问答能力。 large language model chain-of-thought
10 Coal Mining Question Answering with LLMs 提出多轮提示工程框架，提升LLM在煤矿问答中的准确性和相关性 large language model chain-of-thought
11 Unlocking Structured Thinking in Language Models with Cognitive Prompting 提出认知提示方法，利用结构化认知操作提升LLM在复杂问题上的推理能力 large language model chain-of-thought
12 Is Your Paper Being Reviewed by an LLM? Investigating AI Text Detectability in Peer Review 提出新型AI文本检测方法，提升同行评审中LLM代写稿件的识别率 large language model
13 Neutral Residues: Revisiting Adapters for Model Extension 提出中性残差适配器，解决LLM领域迁移中的灾难性遗忘问题 large language model
14 Salient Information Prompting to Steer Content in Prompt-based Abstractive Summarization 提出显著信息提示以优化基于提示的抽象摘要生成 large language model ✅
15 Adaptive Inference-Time Compute: LLMs Can Predict if They Can Do Better, Even Mid-Generation 提出自评估生成方案，LLM可预测自身生成质量并自适应调整计算量，提升推理效率。 large language model
16 LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations 揭示LLM幻觉的内在表征：模型知道的比表现出的更多 large language model
17 Position: LLM Unlearning Benchmarks are Weak Measures of Progress LLM卸载基准测试存在缺陷，无法有效衡量卸载进展 large language model
18 GPT-4o as the Gold Standard: A Scalable and General Purpose Approach to Filter Language Model Pretraining Data 提出SIEVE：一种低成本、可扩展的语言模型预训练数据过滤方法，性能媲美GPT-4o。 large language model
19 UncertaintyRAG: Span-Level Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation UncertaintyRAG：利用跨度不确定性增强长文本RAG，提升模型校准与泛化能力 large language model
20 HiddenGuard: Fine-Grained Safe Generation with Specialized Representation Router HiddenGuard：利用专用表示路由实现大语言模型细粒度安全生成 large language model
21 Hate Personified: Investigating the role of LLMs in content moderation 研究LLM在内容审核中对地域、身份和数值信息的敏感性，揭示其偏见与潜力。 large language model
22 Agents' Room: Narrative Generation through Multi-step Collaboration 提出Agents' Room框架，通过多智能体协作生成高质量叙事故事 large language model
23 Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions 提出两种策略，缓解多智能体LLM交互中存在的隐式性别偏见问题 large language model
24 BrainTransformers: SNN-LLM BrainTransformers：基于脉冲神经网络的大语言模型，提升能效与生物合理性 large language model
25 Revealing the Inherent Instructability of Pre-Trained Language Models 提出响应调优(RT)，揭示预训练语言模型固有的指令理解能力 large language model
26 Better Call SAUL: Fluent and Consistent Language Model Editing with Generation Regularization SAUL：通过生成正则化实现流畅且一致的语言模型编辑 large language model
27 Collective Critics for Creative Story Generation 提出CritiCS框架，通过集体评论机制提升长篇故事生成的创造性和读者参与度。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
28 Grounding Large Language Models In Embodied Environment With Imperfect World Models GLIMO：利用不完美世界模型，提升大语言模型在具身环境中的表现 world model large language model
29 MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions 提出MA-RLHF，通过宏动作提升RLHF在长序列任务中的学习效率与稳定性。 reinforcement learning RLHF large language model ✅
30 Visual Editing with LLM-based Tool Chaining: An Efficient Distillation Approach for Real-Time Applications 提出一种基于LLM工具链的视觉编辑蒸馏方法，用于实时应用。 distillation
31 LLM-Pilot: Characterize and Optimize Performance of your LLM Inference Services LLM-Pilot：表征并优化LLM推理服务的性能，实现硬件高效选择。 predictive model large language model
32 Contextual Document Embeddings 提出上下文文档嵌入方法，提升神经检索中文档表示的质量和泛化性 contrastive learning distillation

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
33 Mixed-Session Conversation with Egocentric Memory 提出MiSC数据集与EMMA模型，解决多轮对话中长期记忆与多方交互问题 egocentric

⬅️ 返回 cs.CL 首页 · 🏠 返回主页