cs.CL（2024-09-02）

📊 共 15 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 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (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) (13 篇)

#	题目	一句话要点	标签	🔗	⭐
1	DiversityMedQA: Assessing Demographic Biases in Medical Diagnosis using Large Language Models	提出DiversityMedQA，评估大型语言模型在医疗诊断中对不同人口统计学信息的偏见	large language model
2	Agentic Society: Merging skeleton from real world and texture from Large Language Model	提出Agentic Society框架，融合真实世界骨架与大语言模型纹理生成虚拟人口。	large language model	✅
3	Large Language Models for Automatic Detection of Sensitive Topics	利用大型语言模型自动检测敏感话题，提升在线社区内容审核效率。	large language model
4	The Compressor-Retriever Architecture for Language Model OS	提出Compressor-Retriever架构，解决LLM OS中的长程上下文管理难题	large language model multimodal	✅
5	Language Models Benefit from Preparation with Elicited Knowledge	提出PREP方法，通过知识引导提升语言模型在问答任务中的性能	instruction following chain-of-thought
6	ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI Systems	提出ComfyBench评估LLM智能体在ComfyUI中自主设计协同AI系统的能力，并提出ComfyAgent框架。	instruction following
7	Membership Inference Attacks Against In-Context Learning	针对上下文学习的成员推理攻击，仅依赖生成文本即可实现高精度隐私泄露。	large language model
8	CHESS: Optimizing LLM Inference via Channel-Wise Thresholding and Selective Sparsification	CHESS：通过通道阈值和选择性稀疏化优化LLM推理	large language model
9	Prompt Compression with Context-Aware Sentence Encoding for Fast and Improved LLM Inference	提出上下文感知提示压缩方法，加速并提升LLM推理性能	large language model	✅
10	NYK-MS: A Well-annotated Multi-modal Metaphor and Sarcasm Understanding Benchmark on Cartoon-Caption Dataset	NYK-MS：一个高质量卡通-文字多模态隐喻与讽刺理解基准数据集	large language model
11	Task-Specific Directions: Definition, Exploration, and Utilization in Parameter Efficient Fine-Tuning	提出LoRA-Dash和LoRA-Init，通过优化任务特定方向提升参数高效微调性能。	large language model
12	DataSculpt: Crafting Data Landscapes for Long-Context LLMs through Multi-Objective Partitioning	DataSculpt：通过多目标划分构建长文本LLM的数据集	large language model
13	User-Specific Dialogue Generation with User Profile-Aware Pre-Training Model and Parameter-Efficient Fine-Tuning	提出用户画像感知的预训练模型与参数高效微调方法，用于生成用户特定对话	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
14	Self-Judge: Selective Instruction Following with Alignment Self-Evaluation	提出Self-Judge框架，通过自评估提升大语言模型指令遵循的可靠性。	distillation large language model instruction following

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

#	题目	一句话要点	标签	🔗	⭐
15	THInC: A Theory-Driven Framework for Computational Humor Detection	提出THInC框架，基于多种幽默理论实现可解释的计算幽默检测	HuMoR large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页