当前位置:首页 > 编程笔记 > 正文
已解决

大模型LLM论文目录

来自网友在路上 172872提问 提问时间:2023-11-05 12:47:32阅读次数: 72

最佳答案 问答题库728位专家为你答疑解惑

持续更新中ing!!!
友情链接:大模型相关资料、基础技术和排行榜

大模型LLM论文目录

标题和时间作者来源简介Artificial General Intelligence: Concept, State of the Art, and Future Prospects,2014GoertzelJournal of Artificial General Intelligence14年的一篇AGI综述,里面探讨了AGI的定义、分类和评估方法等,作者貌似现在是AGI大会的编辑了hhTowards artificial general intelligence with hybrid Tianjic chip architecture,2020Pei jingNature2020年的一个讨论实现AGI硬件的论文,其实现了在同一芯片上同时运行MLP-like和SNN神经网络的硬件环境AGI Brain II: The Upgraded Version with Increased Versatility Index,2021Mohammadreza AlidoustAGI20211.提出一个AGI指标,2.用Mamdani模糊推理联想记忆代替原本的神经网络NN表示外显记忆Training language models to follow instructions with human feedback,2022Long Ouyang等人OpenAIInstructGPT,在大型语言模型的基础上引入人工引导和强化学习,大大提升模型性能Yann Lecun: A Path Towards Autonomous Machine Intelligence 自主机器学习和AGI,2022Yann LecunMachine Learning提出了自主智能体的架构和训练范式,论文地址GPT-4原论文详细解读(GPT-4 Technical Report),2023OpenAIOpenAIGPT-4,提出了多模态的大型语言模型,具备一定的常识和认知能力ChatGLM,2023Aohan Zeng,Du等人International Conference on Learning Representations (ICLR)ChatGLM,ChatGLM-6B结合模型量化技术,用户可以在消费级的显卡上进行本地部署(INT4 量化级别下最低只需 6GB 显存)LLaMA: Open and Efficient Foundation Language Models,2023Hugo TouvronpreprintLLaMA 是 Meta AI 发布的包含 7-65B 参数规模的LLM,其中LLaMA-13B 仅以 1/10 规模的参数在多数的 benchmarks 上性能优于 GPT-3(175B)。开源。A Survey of Large Language Models,2023Wayne Xin Zhao,preprint大型语言模型综述,非常详细,格局打开!ChatDB: AUGMENTING LLMS WITH DATABASES AS THEIR SYMBOLIC MEMORY,2023Chenxu HupreprintChatDB清华团队针对大模型LLMs的长期记忆能力进行的改进,提出数据库与大模型结合开源LONGNET: Scaling Transformers to1,000,000,000 Tokens,2023Jiayu DingpreprintLONGNET微软做的针对大模型的长文本学习,长期记忆进行的改进,开源Focused Transformer: Contrastive Training for Context Scaling,2023Szymon TworkowskipreprintLongLlama谷歌DeepMind研究团队提出了一种注意力集中的transformer架构FOTTowards Benchmarking and Improving the Temporal Reasoning Capability of Large Language Models,2023谭清宇,Hwee Tou Ng,邴立东ACL 2023 main conferenceLLM理解时间变迁。达摩院联合NUS提出时间推理数据集以及时间强化的训练范式UnIVAL: Unified Model for Image, Video, Audio and Language Tasks,2023Mustafa ShukorpreprintUnIVAL,该算法不依赖于数据集大小或具有数十亿参数的大模型,仅仅具有约0.25B的参数量,而且将文本、图像、视频和音频这4个多模态任务统一到了一个模型中Graph of Thoughts: Solving Elaborate Problems with Large Language Models,2023Besta Maciejpreprint思维图,将LLM生成的信息建模为任意图,其中信息单位是顶点,边代表顶点之间的依赖关系The Rise and Potential of Large Language Model Based Agents: A Survey,2023Xi Zhi hengpreprintAgent,综述NExT-GPT: Any-to-Any Multimodal LLM,2023Wu ShengqiongpreprintNExT-GPT,多模态大模型,实现任意模态之间的转换。NextGPT整体结构图、模型示意图和使用模型时示意图Toolformer: Language Models Can Teach Themselves to Use Tools,2023Schick TimopreprintToolsformer,GPT与各种工具结合The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision),2023Yang ZhengyuanpreprintGPT-4V测评报告EFFICIENT STREAMING LANGUAGE MODELS WITH ATTENTION SINKS,2023Xiao Guangxuanpreprint流式LLM,无限扩展LLM长度Improving Image Generation with Better Captions,2023Betker JamesOpen AIDaLLE3,作画大师接入chatgpt,论文中文版见这Instruction Tuning for Large Language Models: A Survey,2023Zhang Linfengpreprint思维链综述RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language ModelsWang Zekun Moorepreprint角色扮演大模型A Survey on Multimodal Large Language Models,2023Yin Chaoyoupreprint多模态大模型综述Visual Instruction Tuning,2023Liu Haotianpreprint视觉大模型llava,通过视觉调优,支持基于图片的聊天
查看全文

99%的人还看了

猜你感兴趣

版权申明

本文"大模型LLM论文目录":http://eshow365.cn/6-32706-0.html 内容来自互联网,请自行判断内容的正确性。如有侵权请联系我们,立即删除!