全网国内外总结Prompt&LLM论文，开源数据&模型，AIGC应用（持续更新，收藏查看）
全网国内外总结Prompt&LLM论文，开源数据&模型，AIGC应用（持续更新，收藏查看）
目录顺序如下
- 国内外，垂直领域大模型
- Agent和指令微调等训练框架
- 开源指令，预训练，rlhf，对话，agent训练数据梳理
- AIGC相关应用
- prompt写作指南和5星博客等资源梳理
- Prompt和LLM论文细分方向梳理
Prompt是在自然语言处理（NLP）中，尤其是在使用预训练语言模型时，用来引导模型生成特定类型输出的一段文本或问题。有效的Prompt设计可以显著提高模型的表现，使其更好地适应特定任务。Prompt可以是具体的指令、问题或者示例，它们充当了模型输入的一部分，帮助模型理解所需完成的任务。
大型语言模型，如GPT-3、BERT和T5，是通过在大量文本数据上进行预训练而得到的复杂神经网络模型。这些模型能够捕捉语言的深层次结构和语义，从而在多种NLP任务中取得卓越表现，包括文本生成、翻译、摘要、问答等。LLM通常需要大量的计算资源进行训练，因此对硬件和数据集的规模有较高要求。
开源数据和模型是指那些公开可用、可以自由使用和修改的数据集和预训练模型。开源数据集，如维基百科、Common Crawl和ImageNet，为训练和测试模型提供了丰富的资源。开源模型，如Hugging Face的Transformers库，提供了预训练模型的实现，允许研究者和开发者在现有工作的基础上进行创新和应用开发。
AIGC（人工智能生成内容）指的是利用人工智能技术自动生成内容的应用。这些内容可以是文本、图像、音频或视频。AIGC应用的例子包括：
- 文本生成：如自动写作助手、新闻生成、社交媒体内容创作等。
- 艺术创作：如AI绘画、音乐创作、诗歌生成等。
- 媒体编辑：如视频剪辑、图片修复、语音合成等。
- 游戏开发：自动生成游戏关卡、角色对话和故事情节。
AIGC技术的发展为内容创作带来了革命性的变化，使得个人和企业能够以更低的成本和更高的效率创造新的内容。
Prompt、LLM、开源数据与模型以及AIGC应用共同推动了人工智能在内容生成领域的进步。Prompt技术提高了LLM在特定任务上的表现，开源数据与模型降低了研究和开发的门槛，而AIGC应用则展示了人工智能在创造性任务中的潜力。随着技术的不断进步，我们可以期待未来会有更多创新的AIGC应用出现，进一步改变我们与信息和媒体的互动方式。
学习预热
- 解密Prompt系列1. Tunning-Free Prompt：GPT2 & GPT3 & LAMA & AutoPrompt
- 解密Prompt系列2. 冻结Prompt微调LM： T5 & PET & LM-BFF
- 解密Prompt系列3. 冻结LM微调Prompt: Prefix-tuning & Prompt-tuning & P-tuning
- 解密Prompt系列4. 升级Instruction Tuning：Flan/T0/InstructGPT/TKInstruct
- 解密prompt系列5. APE+SELF=自动化指令集构建代码实现
- 解密Prompt系列6. lora指令微调扣细节-请冷静,1个小时真不够~
- 解密Prompt系列7. 偏好对齐RLHF-OpenAI·DeepMind·Anthropic对比分析
- 解密Prompt系列8. 无需训练让LLM支持超长输入:知识库 & Unlimiformer & PCW & NBCE
- 解密Prompt系列9. 模型复杂推理-思维链基础和进阶玩法
- 解密Prompt系列10. 思维链COT原理探究
- 解密Prompt系列11. 小模型也能COT，先天不足后天补
- 解密Prompt系列12. LLM Agent零微调范式 ReAct & Self Ask
- 解密Prompt系列13. LLM Agent指令微调方案: Toolformer & Gorilla
- 解密Prompt系列14. LLM Agent之搜索应用设计：WebGPT & WebGLM & WebCPM
- 解密Prompt系列15. LLM Agent之数据库应用设计：DIN & C3 & SQL-Palm & BIRD
- 解密Prompt系列16. LLM对齐经验之数据越少越好？LTD & LIMA & AlpaGasus
- 解密Prompt系列17. LLM对齐方案再升级 WizardLM & BackTranslation & SELF-ALIGN
- 解密Prompt系列18. LLM Agent之只有智能体的世界
- 解密Prompt系列19. LLM Agent之数据分析领域的应用：Data-Copilot & InsightPilot
- 解密Prompt系列20. LLM Agent 之再谈RAG的召回多样性优化
- 解密Prompt系列21. LLM Agent之再谈RAG的召回信息密度和质量
- 解密Prompt系列22. LLM Agent之RAG的反思：放弃了压缩还是智能么？
- 解密Prompt系列23.大模型幻觉分类&归因&检测&缓解方案脑图全梳理
- 解密prompt系列24.RLHF新方案之训练策略：SLiC-HF & DPO & RRHF & RSO
- 解密prompt25. RLHF改良方案之样本标注：RLAIF & SALMON
- 解密prompt系列26. 人类思考vs模型思考：抽象和发散思维
一分钟上手系列：https://blog.csdn.net/u014374009/category_12451843.html
LLMS
模型评测
| 榜单 | 结果 |
|---|---|
| AlpacaEval：LLM-based automatic evaluation | 开源模型王者vicuna,openchat, wizardlm |
| Huggingface Open LLM Leaderboard | MMLU只评估开源模型，Falcon夺冠，在Eleuther AI4个评估集上评估的LLM模型榜单,vicuna夺冠 |
| https://opencompass.org.cn/ | 上海人工智能实验室推出的开源榜单 |
| Berkley出品大模型排位赛榜有准中文榜单 | Elo评分机制，GPT4自然是稳居第一，GPT4>Claude>GPT3.5>Vicuna>others |
| CMU开源聊天机器人评测应用 | ChatGPT>Vicuna>others；在对话场景中训练可能很重要 |
| Z-Bench中文真格基金评测 | 国产中文模型的编程可用性还相对较低，大家水平差不太多，两版ChatGLM提升明显 |
| Chain-of-thought评估 | GSM8k, MATH等复杂问题排行榜 |
| InfoQ 大模型综合能力评估 | 面向中文，ChatGPT>文心一言> Claude>星火 |
| ToolBench: 工具调用评估榜单 | 工具微调模型和ChatGPT进行对比，提供评测脚本 |
| AgentBench: 推理决策评估榜单 | 清华联合多高校推出不同任务环境，例如购物，家居，操作系统等场景下模型推理决策能力 |
| FlagEval | 智源出品主观+客观LLM评分榜单 |
| Bird-Bench | 更贴合真实世界应用的超大数据库，需要领域知识的NL2SQL榜单，模型追赶人类尚有时日 |
| kola | 以世界知识为核心的评价基准，包括已知的百科知识和未知的近90天网络发布内容，评价知识记忆，理解，应用和创造能力 |
| CEVAL | 中文知识评估，覆盖52个学科，机器评价主要为多项选择 |
| CMMLU | 67个主题中文知识和推理能力评估，多项选择机器评估 |
| LLMEval3 | 复旦推出的知识问答榜单，涵盖大学作业和考题，题库尽可能来自非互联网避免模型作弊 |
| QuantBench | AI驱动投资的量化榜单 |
一分钟上手系列：https://blog.csdn.net/u014374009/category_12451843.html
国外开源模型
| 模型链接 | 模型描述 |
|---|---|
| OpenSora | 没等来OpenAI却等来了OpenSora这个梗不错哦 |
| GROK | 马斯克开源Grok-1：3140亿参数迄今最大，权重架构全开放 |
| Gemma | 谷歌商场开源模型2B，7B免费商用，开源第一易主了 |
| Mixtral | 法国“openai”开源基于MegaBlocks训练的MOE模型8*7B 32K |
| Mistral7B | 法国“openai”开源Mistral，超过llama2当前最好7B模型 |
| Dolphin-2.2.1-Mistral-7B | 基于Mistral7B使用dolphin数据集微调 |
| LLama2 | Open Meta带着可商用开源的羊驼2模型来了~ |
| LLaMA | Meta开源指令微调LLM，规模70 亿到 650 亿不等 |
| WizardLM | 微软新发布13B，登顶AlpacaEval开源模型Top3，使用ChatGPT对指令进行复杂度进化微调LLama2 |
| Falcon | Falcon由阿联酋技术研究所在超高质量1万亿Token上训练得到1B，7B，40B开源，免费商用！土豪们表示钱什么的格局小了 |
| Vicuna | Alpaca前成员等开源以LLama13B为基础使用ShareGPT指令微调的模型，提出了用GPT4来评测模型效果 |
| OpenChat | 80k ShareGPT对话微调LLama-2 13B开源模型中的战斗机 |
| Guanaco | LLama 7B基座，在alpaca52K数据上加入534K多语言指令数据微调 |
| MPT | MosaicML开源的预训练+指令微调的新模型，可商用，支持84k tokens超长输入 |
| RedPajama | RedPajama项目既开源预训练数据后开源3B，7B的预训练+指令微调模型 |
| koala | 使用alpaca，HC3等开源指令集+ ShareGPT等ChatGPT数据微调llama，在榜单上排名较高 |
| ChatLLaMA | 基于RLHF微调了LLaMA |
| Alpaca | 斯坦福开源的使用52k数据在7B的LLaMA上微调得到， |
| Alpaca-lora | LORA微调的LLaMA |
| Dromedary | IBM self-aligned model with the LLaMA base |
| ColossalChat | HPC-AI Tech开源的Llama+RLHF微调 |
| MiniGPT4 | Vicuna+BLIP2 文本视觉融合 |
| StackLLama | LLama使用Stackexchange数据+SFT+RL |
| Cerebras | Cerebras开源了1亿到130亿的7个模型，从预训练数据到参数全开源 |
| Dolly-v2 | 可商用 7b指令微调开源模型在GPT-J-6B上微调 |
| OpenChatKit | openai研究员打造GPT-NoX-20B微调+6B审核模型过滤 |
| MetaLM | 微软开源的大规模自监督预训练模型 |
| Amazon Titan | 亚马逊在aws上增加自家大模型 |
| OPT-IML | Meta复刻GPT3，up to 175B, 不过效果并不及GPT3 |
| Bloom | BigScience出品，规模最大176B |
| BloomZ | BigScience出品, 基于Bloom微调 |
| Galacia | 和Bloom相似，更针对科研领域训练的模型 |
| T0 | BigScience出品，3B~11B的在T5进行指令微调的模型 |
| EXLLama | Python/C++/CUDA implementation of Llama for use with 4-bit GPTQ weight |
| LongChat | llama-13b使用condensing rotary embedding technique微调的长文本模型 |
| MPT-30B | MosaicML开源的在8Ktoken上训练的大模型 |
国内开源模型
| 模型链接 | 模型描述 |
|---|---|
| Baichuan2 | 百川第二代，提供了7B/13B Base和chat的版本 |
| Baichuan | 百川智能开源7B大模型可商用免费 |
| ziya2 | 基于Llama2训练的ziya2它终于训练完了 |
| ziya | IDEA研究院在7B/13B llama上继续预训练+SFT+RM+PPO+HFTT+COHFT+RBRS |
| Qwen1.5 | 通义千问升级1.5，支持32K上文 |
| Qwen1-7B+14B+70B | 阿里开源，可商用，通义千问7B,14B,70B Base和chat模型 |
| InternLM2 7B+20B | 商汤的书生模型2支持200K |
| Orion-14B-LongChat | 猎户星空多语言模型支持320K |
| ChatGLM3 | ChatGLM3发布，支持工具调用等更多功能，不过泛化性有待评估 |
| ChatGLM2 | 32K长文本，FlashAttention+Multi-Query Attenion的显存优化，更强推理能力，哈哈不过很多简单问题也硬要COT，中英平行能力似乎略有下降的ChatGLM2，但是免费商用！ |
| ChatGLM | 清华开源的、支持中英双语的对话语言模型，使用了代码训练，指令微调和RLHF。chatglm2支持超长文本，可免费商用啦！ |
| Yuan-2.0 | 浪潮发布Yuan2.0 2B，51B，102B |
| YI-200K | 元一智能开源超长200K的6B，34B模型 |
| YI | 元一智能开源34B，6B模型 |
| XVERSE-256K | 元象发布13B免费商用大模型，虽然很长但是 |
| XVERSE | 元象发布13B免费商用大模型 |
| DeepSeek-MOE | 深度求索发布的DeepSeekMoE 16B Base和caht模型 |
| DeepSeek | 深度求索发布的7B，67B大模型 |
| LLama2-chinese | 没等太久中文预训练微调后的llama2它来了~ |
| YuLan-chat2 | 高瓴人工智能基于Llama-2中英双语继续预训练+指令微调/对话微调 |
| BlueLM | Vivo人工智能实验室开源大模型 |
| zephyr-7B | HuggingFace 团队基于 UltraChat 和 UltraFeedback 训练了 Zephyr-7B 模型 |
| XWin-LM | llama2 + SFT + RLHF |
| Skywork | 昆仑万维集团·天工团队开源13B大模型可商用 |
| Chinese-LLaMA-Alpaca | 哈工大中文指令微调的LLaMA |
| Moss | 为复旦正名！开源了预训练，指令微调的全部数据和模型。可商用 |
| InternLM | 书生浦语在过万亿 token 数据上训练的多语千亿参数基座模型 |
| Aquila2 | 智源更新Aquila2模型系列包括全新34B |
| Aquila | 智源开源7B大模型可商用免费 |
| UltraLM系列 | 面壁智能开源UltraLM13B，奖励模型UltraRM，和批评模型UltraCM |
| PandaLLM | LLAMA2上中文wiki继续预训练+COIG指令微调 |
| XVERSE | 据说中文超越llama2的元象开源模型13B模型 |
| BiLLa | LLama词表·扩充预训练+预训练和任务1比1混合SFT+指令样本SFT三阶段训练 |
| Phoenix | 港中文开源凤凰和奇美拉LLM，Bloom基座，40+语言支持 |
| Wombat-7B | 达摩院开源无需强化学习使用RRHF对齐的语言模型, alpaca基座 |
| TigerBot | 虎博开源了7B 180B的模型以及预训练和微调语料 |
| Luotuo | 中文指令微调的LLaMA，和ChatGLM |
| OpenBuddy | Llama 多语言对话微调模型 |
| Chinese Vincuna | LLama 7B基座，使用Belle+Guanaco数据训练 |
| Linly | Llama 7B基座，使用belle+guanaco+pclue+firefly+CSL+newscommentary等7个指令微调数据集训练 |
| Firefly | 中文2.6B模型，提升模型中文写作，古文能力，待开源全部训练代码，当前只有模型 |
| Baize | 使用100k self-chat对话数据微调的LLama |
| BELLE | 使用ChatGPT生成数据对开源模型进行中文优化 |
| Chatyuan | chatgpt出来后最早的国内开源对话模型，T5架构是下面PromptCLUE的衍生模型 |
| PromptCLUE | 多任务Prompt语言模型 |
| PLUG | 阿里达摩院发布的大模型，提交申请会给下载链接 |
| CPM2.0 | 智源发布CPM2.0 |
| GLM | 清华发布的中英双语130B预训练模型 |
| BayLing | 基于LLama7B/13B，增强的语言对齐的英语/中文大语言模型 |
一分钟上手系列：https://blog.csdn.net/u014374009/category_12451843.html
国内外免费试用的大模型应用
| 模型链接 | 模型描述 |
|---|---|
| PPLX-7B/70B | Perplexity.ai的Playground支持他们自家的PPLX模型和众多SOTA大模型，Gemma也支持了 |
| kimi Chat | Moonshot超长文本LLM 可输入20W上文, 文档总结无敌 |
| 讯飞星火 | 科大讯飞 |
| 文心一言 | 百度 |
| 通义千问 | 阿里 |
| 百川 | 百川 |
| ChatGLM | 智谱轻言 |
| DeepSeek | 深度求索 |
| 360智脑 | 360 |
| 悟空 | 字节跳动 |
垂直领域模型&进展
| 领域 | 模型链接 | 模型描述 |
|---|---|---|
| 医疗 | MedGPT | 医联发布的 |
| 医疗 | MedPalm | Google在Faln-PaLM的基础上通过多种类型的医疗QA数据进行prompt-tuning指令微调得到，同时构建了MultiMedQA |
| 医疗 | ChatDoctor | 110K真实医患对话样本+5KChatGPT生成数据进行指令微调 |
| 医疗 | Huatuo Med-ChatGLM | 医学知识图谱和chatgpt构建中文医学指令数据集+医学文献和chatgpt构建多轮问答数据 |
| 医疗 | Chinese-vicuna-med | Chinese-vicuna在cMedQA2数据上微调 |
| 医疗 | OpenBioMed | 清华AIR开源轻量版BioMedGPT, 知识图谱&20+生物研究领域多模态预训练模型 |
| 医疗 | DoctorGLM | ChatDoctor+MedDialog+CMD 多轮对话+单轮指令样本微调GLM |
| 医疗 | MedicalGPT-zh | 自建的医学数据库ChatGPT生成QA+16个情境下SELF构建情景对话 |
| 医疗 | PMC-LLaMA | 医疗论文微调Llama |
| 医疗 | PULSE | Bloom微调+继续预训练 |
| 医疗 | NHS-LLM | Chatgpt生成的医疗问答，对话，微调模型 |
| 医疗 | 神农医疗大模型 | 以中医知识图谱的实体为中心生成的中医知识指令数据集11w+，微调LLama-7B |
| 医疗 | 岐黄问道大模型 | 3个子模型构成，已确诊疾病的临床治疗模型+基于症状的临床诊疗模型+中医养生条理模型，看起来是要ToB落地 |
| 医疗 | Zhongjing | 基于Ziya-LLama+医疗预训练+SFT+RLHF的中文医学大模型 |
| 医疗 | MeChat | 心理咨询领域，通过chatgpt改写多轮对话56k |
| 医疗 | SoulChat | 心理咨询领域中文长文本指令与多轮共情对话数据联合指令微调 ChatGLM-6B |
| 医疗 | MindChat | MindChat-Baichuan-13B,Qwen-7B,MindChat-InternLM-7B使用不同基座在模型安全，共情，人类价值观对其上进行了强化 |
| 医疗 | DISC-MedLLM | 疾病知识图谱构建QA对+QA对转化成单论对话+真实世界数据重构+人类偏好数据筛选，SFT微调baichuan |
| 法律 | LawGPT-zh | 利用ChatGPT清洗CrimeKgAssitant数据集得到52k单轮问答+我们根据中华人民共和国法律手册上最核心的9k法律条文，利用ChatGPT联想生成具体的情景问答+知识问答使用ChatGPT基于文本构建QA对 |
| 法律 | LawGPT | 基于llama+扩充词表二次预训练+基于法律条款构建QA指令微调 |
| 法律 | Lawyer Llama | 法律指令微调数据集：咨询+法律考试+对话进行指令微调 |
| 法律 | LexiLaw | 法律指令微调数据集：问答+书籍概念解释，法条内容进行指令微调 |
| 法律 | ChatLaw | 北大推出的法律大模型，应用形式很新颖类似频道内流一切功能皆融合在对话形式内 |
| 法律 | 录问模型 | 在baichuan基础上40G二次预训练+100K指令微调，在知识库构建上采用了Emb+意图+关键词联想结合的方案 |
| 金融 | FinChat.io | 使用最新的财务数据，电话会议记录，季度和年度报告，投资书籍等进行训练 |
| 金融 | OpenGPT | 领域LLM指令样本生成+微调框架 |
| 金融 | 乾元BigBang金融2亿模型 | 金融领域预训练+任务微调 |
| 金融 | 度小满千亿金融大模型 | 在Bloom-176B的基础上进行金融+中文预训练和微调 |
| 金融 | bondGPT | GPT4在细分债券市场的应用开放申请中 |
| 金融 | IndexGPT | JPMorgan在研的生成式投资顾问 |
| 金融 | 恒生LightGPT | 金融领域继续预训练+插件化设计 |
| 金融 | 知彼阿尔法 | 企查查商查大模型 |
| 金融 | AlphaBox | 熵简科技发布大模型金融应用，多文档问答+会议转录+文档编辑 |
| 金融 | 曹植 | 达观发布金融大模型融合data2text等金融任务，赋能报告写作 |
| 金融 | 聚宝盆 | 基于 LLaMA 系基模型经过中文金融知识指令精调/指令微调(Instruct-tuning) 的微调模型 |
| 金融 | PIXIU | 整理了多个金融任务数据集加入了时间序列数据进行指令微调 |
| 金融 | ChatFund | 韭圈儿发布的第一个基金大模型，看起来是做了多任务指令微调，和APP已有的数据功能进行了全方位的打通，从选基，到持仓分析等等 |
| 金融 | FinGPT | 金融传统任务微调 or chatgpt生成金融工具调用 |
| 金融 | CFGPT | 金融预训练+指令微调+RAG等检索任务增强 |
| 金融 | 况客FOF智能投顾 | 基金大模型应用，基金投顾，支持nl2sql类的数据查询，和基金信息对比查询等 |
| 金融 | DISC-FinLLM | 复旦发布多微调模型组合金融系统，包括金融知识问答,金融NLP任务，金融计算，金融检索问答 |
| 金融 | InvestLM | CFA考试，SEC， StackExchange投资问题等构建的金融指令微调LLaMA-65+ |
| 金融 | HithinkGPT | 同花顺发布金融大模型问财，覆盖查询，分析，对比，解读，预测等多个问题领域 |
| 金融 | 无涯Infinity | 星环科技发布的金融大模型 |
| 金融 | 妙想 | 东方财富自研金融大模型开放试用 |
| 金融 | DeepMoney | 基于yi-34b-200k使用金融研报进行微调 |
| 编程 | Starcoder | 80种编程语言+Issue+Commit训练得到的编程大模型 |
| 编程 | ChatSQL | 基于ChatGLM实现NL2sql |
| 编程 | codegeex | 13B预训练+微调多语言变成大模型 |
| 编程 | codegeex2 | Chatglm2的基础上CodeGeeX2-6B 进一步经过了 600B 代码数据预训练 |
| 编程 | stabelcode | 560B token多语言预训练+ 120,000 个 Alpaca指令对齐 |
| 编程 | SQLCoder | 在StarCoder的基础上微调15B超越gpt3.5 |
| 数学 | MathGPT | 是好未来自主研发的，面向全球数学爱好者和科研机构，以解题和讲题算法为核心的大模型。 |
| 数学 | MammoTH | 通过COT+POT构建了MathInstruct数据集微调llama在OOD数据集上超越了WizardLM |
| 数学 | MetaMath | 模型逆向思维解决数学问题，构建了新的MetaMathQA微调llama2 |
| 交通 | TransGPT | LLama-7B+34.6万领域预训练+5.8万条领域指令对话微调（来自文档问答） |
| 交通 | TrafficGPT | ChatGPT+Prompt实现规划，调用交通流量领域专业TFM模型，TFM负责数据分析，任务执行，可视化等操作 |
| 科技 | Mozi | 红睡衣预训练+论文QA数据集 + ChatGPT扩充科研对话数据 |
| 天文 | StarGLM | 天文知识指令微调，项目进行中后期考虑天文二次预训练+KG |
| 写作 | 阅文-网文大模型介绍 | 签约作者内测中，主打的内容为打斗场景，剧情切换，环境描写，人设，世界观等辅助片段的生成 |
| 写作 | MediaGPT | LLama-7B扩充词表+指令微调，指令来自国内媒体专家给出的在新闻创作上的80个子任务 |
| 电商 | EcomGPT | 电商领域任务指令微调大模型，指令样本250万，基座模型是Bloomz |
| 植物科学 | PLLaMa | 基于Llama使用植物科学领域学术论文继续预训练+sft扩展的领域模型 |
| 评估 | Auto-J | 上交开源了价值评估对齐13B模型 |
| 评估 | JudgeLM | 智源开源了 JudgeLM 的裁判模型，可以高效准确地评判各类大模型 |
| 评估 | CritiqueLLM | 智谱AI发布评分模型CritiqueLLM,支持含参考文本/无参考文本的评估打分 |
新时代RAG Embedding模型
| 模型链接 | 模型描述 |
|---|---|
| Jina-Cobert | Jian AI开源中英德，8192 Token长文本Embedding |
| BGE-M3 | 智源开源多语言，稀疏+稠密表征，8192 Token长文本Embedding |
| BCE | 网易开源更适配RAG任务的Embedding模型 |
Tool and Library
推理框架
| 工具描述 | 链接 |
|---|---|
| FlexFlow：模型部署推理框架 | https://github.com/flexflow/FlexFlow |
| Medusa：针对采样解码的推理加速框架，可以和其他策略结合 | https://github.com/FasterDecoding/Medusa |
| FlexGen: LLM推理 CPU Offload计算架构 | https://github.com/FMInference/FlexGen |
| VLLM：超高速推理框架Vicuna，Arena背后的无名英雄，比HF快24倍，支持很多基座模型 | https://github.com/vllm-project/vllm |
| Streamingllm: 新注意力池Attention方案，无需微调拓展模型推理长度，同时为推理提速 | https://github.com/mit-han-lab/streaming-llm |
| llama2.c: llama2 纯C语言的推理框架 | https://github.com/karpathy/llama2.c |
指令微调，预训练，rlhf框架
| 工具描述 | 链接 |
|---|---|
| LoRA：Low-Rank指令微调方案 | https://github.com/tloen/alpaca-lora |
| peft：parameter-efficient prompt tunnging工具集 | https://github.com/huggingface/peft |
| RL4LMs：AllenAI的RL工具 | https://github.com/allenai/RL4LMs |
| RLLTE：港大，大疆等联合开源RLLTE开源学习框架 | https://github.com/RLE-Foundation/rllte |
| trl：基于Transformer的强化训练框架 | https://github.com/lvwerra/trl |
| trlx：分布式训练trl | https://github.com/CarperAI/trlx |
| 北大开源河狸项目可复现RLHF，支持多数LLM，提供RLHF数据 | https://github.com/PKU-Alignment/safe-rlhf |
| RL4LMs：AllenAI的RL工具 | https://github.com/allenai/RL4LMs |
| LMFlow：港科大实验室开源的大模型微调框架，支持以上多数开源模型的指令微调和RLHF | https://github.com/OptimalScale/LMFlow |
| hugNLP:基于Huggingface开发继承Prompt技术，预训练和是指输入等多种方案 | https://github.com/wjn1996/HugNLP |
| Deepspeed：针对RL训练和推理的整合优化 | https://github.com/microsoft/DeepSpeed |
| Uerpy:预训练框架支持lm,mlm,unilm等 | https://github.com/dbiir/UER-py |
| TecentPretrain: Uerpy的重构版本支持llama预训练 | https://github.com/Tencent/TencentPretrain/tree/main |
| lamini: 整合指令数据生成，SFT，RLHF的工具库 | https://github.com/lamini-ai/lamini/ |
| Chain-of-thought-hub：模型推理能力评估平台 | https://github.com/FranxYao/chain-of-thought-hub |
| EasyEdit：浙大开源支持多种模型，多种方案的模型知识精准编辑器 | https://github.com/zjunlp/EasyEdit |
| OpenDelta：集成了各种增量微调方案的开源实现 | https://github.com/thunlp/OpenDelta |
| Megablocks：MOE训练框架 | https://github.com/stanford-futuredata/megablocks |
| Tutel：MOE训练框架 | https://github.com/microsoft/tutel |
| TradingGym：参考openai gym的股票交易强化学习模拟器 | https://github.com/astrologos/tradinggym |
| LongLora: 长文本微调框架 | https://github.com/dvlab-research/LongLoRA |
| LlamaGym：在线RL微调框架 | https://github.com/KhoomeiK/LlamaGym |
Auto/Multi Agent
| 工具描述 | 链接 |
|---|---|
| AutoGen：微软开源多Agent顶层框架 | https://github.com/microsoft/autogen |
| CrewAI: 比chatDev流程定义更灵活的多智能体框架 | https://github.com/joaomdmoura/CrewAI |
| ChatDev: 面壁智能开源多智能体协作的虚拟软件公司 | https://github.com/OpenBMB/ChatDev |
| Generative Agents:斯坦福AI小镇的开源代码 | https://github.com/joonspk-research/generative_agents |
| BabyAGI：自执行LLM Agent | https://github.com/yoheinakajima/babyagi |
| AutoGPT：自执行LLM Agent | https://github.com/Torantulino/Auto-GPT |
| AutoGPT-Plugins：提供众多Auo-GPT官方和第三方的插件 | https://github.com/Significant-Gravitas/Auto-GPT-Plugins |
| XAgent: 面壁智能开源双循环AutoGPT | https://github.com/OpenBMB/XAgent |
| MetaGPT: 覆盖软件公司全生命流程，例如产品经理等各个职业的AutoGPT | https://github.com/geekan/MetaGPT |
| ResearchGPT: 论文写作领域的AutoGPT，融合论文拆解+网络爬虫 | https://github.com/assafelovic/gpt-researcher |
| MiniAGI：自执行LLM Agent | https://github.com/muellerberndt/mini-agi |
| AL Legion： 自执行LLM Agent | https://github.com/eumemic/ai-legion |
| AgentVerse：多模型交互环境 | https://github.com/OpenBMB/AgentVerse |
| AgentSims: 给定一个社会环境，评估LLM作为智能体的预定任务目标完成能力的沙盒环境 | https://github.com/py499372727/AgentSims/ |
| GPTRPG：RPG环境 AI Agent游戏化 | https://github.com/dzoba/gptrpg |
| GPTeam：多智能体交互 | https://github.com/101dotxyz/GPTeam |
| GPTEngineer：自动工具构建和代码生成 | https://github.com/AntonOsika/gpt-engineer |
| WorkGPT：类似AutoGPT | https://github.com/team-openpm/workgpt |
| AI-Town: 虚拟世界模拟器 | https://github.com/a16z-infra/ai-town |
| webarena:网络拟真环境，可用于自主智能体的测试，支持在线购物，论坛，代码仓库etc | https://github.com/web-arena-x/webarena |
| MiniWoB++：100+web交互操作的拟真环境 | https://github.com/Farama-Foundation/miniwob-plusplus |
Agent工具框架类
| 工具描述 | 链接 |
|---|---|
| OpenAgents: 开源版ChatGPT-Plus搭建框架 | https://github.com/xlang-ai/OpenAgents |
| langchain：LLM Agent框架 | https://github.com/hwchase17/langchain |
| llama index：LLM Agent框架 | https://github.com/jerryjliu/llama_index |
| Langroid: LLM Agent框架 | https://github.com/langroid/langroid |
| Ragas: 评估检索增强LLM效果的框架，基于大模型prompt评估事实性，召回相关性，召回内容质量，回答相关性等 | https://github.com/explodinggradients/ragas#fire-quickstart |
| fastRAG：检索框架，包括多索引检索，KG构建等基础功能 | https://github.com/IntelLabs/fastRAG/tree/main |
| langflow：把langchain等agent组件做成了可拖拽式的UI | https://github.com/logspace-ai/langflow |
| PhiData：把工具调用抽象成function call的Agent框架 | https://github.com/phidatahq/phidata |
| Haystack: LLM Agent 框架，pipeline的设计模式个人感觉比langchain更灵活更简洁 | https://github.com/deepset-ai/haystack |
| EdgeChain: 通过Jsonnet配置文件实现LLM Agent | https://github.com/arakoodev/EdgeChains/tree/main |
| semantic-kernel：整合大模型和编程语言的SDK | https://github.com/microsoft/semantic-kernel |
| BMTTools: 清华出品多工具调用开源库，提供微调数据和评估ToolBench | https://github.com/OpenBMB/BMTools |
| Jarvis: 大模型调用小模型框架，给小模型一个未来！ | https://github.com/search?q=jarvis |
| LLM-ToolMaker:让LLM自己制造Agent | https://github.com/ctlllll/LLM-ToolMaker |
| Gorilla: LLM调用大量API | https://github.com/ShishirPatil/gorilla |
| wenda:闻达小模型整合搜索用于知识融入 | https://github.com/l15y/wenda |
| Alexandria: 从Arix论文开始把整个互联网变成向量索引，可以免费下载 | https://alex.macrocosm.so/download |
| RapidAPI: 统一这个世界的所有API，最大API Hub，有调用成功率，latency等，是真爱！ | https://rapidapi.com/hub |
| Open-Interpreter：命令行聊天框架 | https://github.com/KillianLucas/open-interpreter |
| AnythingLLM: langchain推出的支持本地部署开源模型的框架 | https://github.com/Mintplex-Labs/anything-llm |
| PromptFlow：微软推出的大模型应用框架 | https://github.com/microsoft/promptflow |
| Coze:字节跳动推出的个性化Agent定制应用支持多个大模型丰富插件集使用 | https://www.coze.com/username?redirect=/explore |
| Anakin：和Coze类似的Agent定制应用，插件支持较少但workflow使用起来更简洁 | https://app.anakin.ai/discover |
| TaskingAI：API-Oriented的类似langchain的大模型应用框架 | https://www.tasking.ai/ |
其他垂直领域Agent
| 工具描述 | 链接 |
|---|---|
| Deep-KE：基于LLM对数据进行智能解析实现知识抽取 | https://github.com/zjunlp/DeepKE |
| IncarnaMind：多文档RAG方案，动态chunking的方案可以借鉴 | https://github.com/junruxiong/IncarnaMind |
| Vectra：平台化的LLM Agent搭建方案，从索引构建，内容召回排序，到事实检查的LLM生成 | https://vectara.com/tour-vectara/ |
| Data-Copilot：时间序列等结构化数据分析领域的Agent解决方案 | https://github.com/zwq2018/Data-Copilot |
| DB-GPT: 以数据库为基础的GPT实验项目，使用本地化的GPT大模型与您的数据和环境进行交互 | https://db-gpt.readthedocs.io/projects/db-gpt-docs-zh-cn/zh_CN/latest/index.html |
| guardrails：降低模型幻觉的python框架，promp模板+validation+修正 | https://github.com/shreyar/guardrails |
| guidance：微软新开源框架，同样是降低模型幻觉的框架，prompt+chain的升级版加入逐步生成和思维链路 | https://github.com/guidance-ai/guidance |
| SolidGPT: 上传个人数据，通过命令交互创建项目PRD等 | https://github.com/AI-Citizen/SolidGPT |
| HR-Agent: 类似HR和员工交互，支持多工具调用 | https://github.com/stepanogil/autonomous-hr-chatbot |
| BambooAI：数据分析Agent | https://github.com/pgalko/BambooAI |
| AlphaCodium：通过Flow Engineering完成代码任务 | https://github.com/Codium-ai/AlphaCodium |
Training Data
| 数据类型 | 数据描述 | 数据链接 |
|---|---|---|
| 指令微调 | self-instruct，GPT3自动生成&过滤得到指令集 | https://github.com/yizhongw/self-instruct |
| 指令微调 | Standford Alpaca：52K text-davinci-003生成的self-instruct指令数据集 | https://github.com/tatsu-lab/stanford_alpaca |
| 指令微调 | GPT4-for-LLM 中文+英文+对比指令 | https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM |
| 指令微调 | GPTTeacher更多样的通用指令，角色扮演和代码指令 | https://github.com/teknium1/GPTeacher/tree/main |
| 指令微调 | 中文翻译Alpaca还有一些其他指令数据集 | https://github.com/hikariming/alpaca_chinese_dataset https://github.com/carbonz0/alpaca-chinese-dataset |
| 指令微调 | alpaca指令GPT4生成，和以上几版对比显著质量更高，回复更长 | https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM/tree/main |
| 指令微调 | Guanaco数据：对Alphca指令重写后以不同语言生成总共534K，有对话和非对话类型，还有补充的QA生成样本 | https://huggingface.co/datasets/JosephusCheung/GuanacoDataset |
| 指令微调 | OIG中文指令包括翻译alpaca+natural+unnatural，多轮对话，考试，leetcode指令 | https://github.com/BAAI-Zlab/COIG |
| 指令微调 | Vicuna训练使用的样本，用API获取了sharegpt上用户和chatgpt对话历史，部分网友整理到了HF | https://github.com/domeccleston/sharegpt https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/tree/main |
| 指令微调 | HC3指令数据中英文，包括金融，开放QA，百科，DBQA，医学等包含人工回复 | https://huggingface.co/datasets/Hello-SimpleAI/HC3-Chinese/tree/main |
| 指令微调 | MOSS开源的SFT数据包含使用plugin的对话数据 | https://huggingface.co/datasets/Hello-SimpleAI/HC3-Chinese/tree/main |
| 指令微调 | InstructWild数据：用四处爬取的chatgpt指令作为种子self-instruct扩充生成，中英双语 | https://github.com/XueFuzhao/InstructionWild/tree/main/data |
| 指令微调 | BELLE100万指令数据，参考Alpaca用ChatGPT生成，有数学，多轮对话，校色对话等等 | https://github.com/LianjiaTech/BELLE |
| 指令微调 | PromptCLUE多任务提示数据集：模板构建，只包含标准NLP任务 | https://github.com/CLUEbenchmark/pCLUE |
| 指令微调 | TK-Instruct微调用的指令数据集, 全人工标注1600+NLP任务 | https://instructions.apps.allenai.org/ |
| 指令微调 | T0微调用的指令数据集（P3） | https://huggingface.co/datasets/bigscience/P3 |
| 指令微调 | p3衍生的46种多语言数据集（xmtf） | https://github.com/bigscience-workshop/xmtf |
| 指令微调 | Unnatural Instruction使用GPT3生成后改写得到240k | https://github.com/orhonovich/unnatural-instructions |
| 指令微调 | alpaca COT对多个数据源进行了清理并统一格式放到的了HF, 重点是人工整理的COT数据 | https://github.com/PhoebusSi/Alpaca-CoT |
| 指令微调 | 人工编写包含23种常见的中文NLP任务的指令数据，中文写作方向 | https://github.com/yangjianxin1/Firefly |
| 指令微调 | Amazon COT指令样本包括各类QA，bigbench，math等 | https://github.com/amazon-science/auto-cot |
| 指令微调 | CSL包含 396,209 篇中文核心期刊论文元信息 （标题、摘要、关键词、学科、门类）可做预训练可构建NLP指令任务 | https://github.com/ydli-ai/CSL |
| 指令微调 | alpaca code 20K代码指令数据 | https://github.com/sahil280114/codealpaca#data-release |
| 指令微调 | GPT4Tools 71K GPT4指令样本 | https://github.com/StevenGrove/GPT4Tools |
| 指令微调 | GPT4指令+角色扮演+代码指令 | https://github.com/teknium1/GPTeacher |
| 指令微调 | Mol-Instructions 2043K 分子+蛋白质+生物分子文本指令，覆盖分子设计、蛋白质功能预测、蛋白质设计等任务 | https://github.com/zjunlp/Mol-Instructions |
| 数学 | 腾讯人工智能实验室发布网上爬取的数学问题APE210k | https://github.com/Chenny0808/ape210k |
| 数学 | 猿辅导 AI Lab开源小学应用题Math23K | https://github.com/SCNU203/Math23k/tree/main |
| 数学 | grade school math把OpenAI的高中数学题有改造成指令样本有2-8步推理过程 | https://huggingface.co/datasets/qwedsacf/grade-school-math-instructions |
| 数学 | 数学问答数据集有推理过程和多项选择 | https://huggingface.co/datasets/math_qa/viewer/default/test?row=2 |
| 数学 | AMC竞赛数学题 | https://huggingface.co/datasets/competition_math |
| 数学 | 线性代数等纯数学计算题 | https://huggingface.co/datasets/math_dataset |
| 代码 | APPS从不同的开放访问编码网站Codeforces、Kattis 等收集的问题 | https://opendatalab.org.cn/APPS |
| 代码 | Lyra代码由带有嵌入式 SQL 的 Python 代码组成，经过仔细注释的数据库操作程序，配有中文评论和英文评论。 | https://opendatalab.org.cn/Lyra |
| 代码 | Conala来自StackOverflow问题,手动注释3k，英文 | https://opendatalab.org.cn/CoNaLa/download |
| 代码 | code-alpaca ChatGPT生成20K代码指令样本 | https://github.com/sahil280114/codealpaca.git |
| 代码 | 32K, 四种不同类型、不同难度的代码相关中文对话数据，有大模型生成， | https://github.com/zxx000728/CodeGPT |
| 对话 | LAION 策划的开放指令通用数据集中手动选择的组件子集 已开源40M 3万个,100M在路上 | https://github.com/LAION-AI/Open-Instruction-Generalist |
| 对话 | Baize基于Chat GPT构建的self-chat数据 | https://github.com/project-baize/baize-chatbot/tree/main/data |
| 对话 | FaceBook开源BlenderBot训练对话数据~6K | https://huggingface.co/datasets/blended_skill_talk |
| 对话 | AllenAI开源38.5万个对话高质量数据集SODA | https://realtoxicityprompts.apps.allenai.org/ |
| 对话 | InstructDial在单一对话任务类型上进行指令微调 | https://github.com/prakharguptaz/Instructdial |
| 对话 | Ultra Chat 两个独立的 ChatGPT Turbo API 进行对话，从而生成多轮对话数据 | https://github.com/thunlp/UltraChat |
| 对话 | Awesome Open-domain Dialogue Models提供多个开放域对话数据 | https://github.com/cingtiye/Awesome-Open-domain-Dialogue-Models#%E4%B8%AD%E6%96%87%E5%BC%80%E6%94%BE%E5%9F%9F%E5%AF%B9%E8%AF%9D%E6%95%B0%E6%8D%AE%E9%9B%86 |
| 对话 | Salesforce开源超全DialogStudio | https://github.com/salesforce/DialogStudio |
| 对话 | 基于事实Reference的多轮问答中文数据，已开源5万，之后会开源更多 | https://github.com/sufengniu/RefGPT |
| RLFH | 北大河狸开源RLHF数据集10K，1M需要申请 | https://huggingface.co/datasets/PKU-Alignment/PKU-SafeRLHF-10K |
| RLHF | Anthropic hh-rlhf数据集 | https://huggingface.co/datasets/Anthropic/hh-rlhf |
| RLHF | Stack-exchange上问题对应多个答案，每个答案有打分 | https://huggingface.co/datasets/HuggingFaceH4/stack-exchange-preferences/tree/main |
| RLHF | Facebook Bot Adversarial Dialogues数据集5K | https://github.com/facebookresearch/ParlAI |
| RLHF | AllenAI Real Toxicity prompts | https://github.com/facebookresearch/ParlAI |
| RLHF | OpenAssistant Conversations 160K消息，13500人工生成, 英文为主 | https://huggingface.co/datasets/OpenAssistant/oasst1 |
| RLHF | 知乎问答偏好数据集 | https://huggingface.co/datasets/liyucheng/zhihu_rlhf_3k |
| RLHF | hh-rlhf中文翻译偏好数据 | https://huggingface.co/datasets/liswei/rm-static-zhTW |
| RLHF | 面壁智能开源大规模偏好数据，基于64Kprompt使用不同模型生成4个回答使用GPT-4评估 | https://github.com/OpenBMB/UltraFeedback |
| 评估集 | BigBench(Beyond the Imitation Game Benchmark) | https://github.com/google/BIG-bench |
| 评估集 | Complex QA：用于ChatGPT的评测指令集 | https://github.com/tan92hl/Complex-Question-Answering-Evaluation-of-ChatGPT |
| 评估集 | Langchain开源评估数据集 | https://huggingface.co/LangChainDatasets |
| 评估集 | 2010-2022年全国高考卷的题目 | https://github.com/OpenLMLab/GAOKAO-Bench |
| 评估集 | 中文通用大模型综合性评测基准SuperCLUE | https://github.com/CLUEbenchmark/SuperCLUE |
| 英文预训练 | RedPajama开源的复刻llama的预训练数据集，1.21万亿Token | https://github.com/togethercomputer/RedPajama-Data |
| 英文预训练 | Cerebras基于RedPajama进行清洗去重后得到的高质量数据集, 6270亿Token | https://huggingface.co/datasets/cerebras/SlimPajama-627B/tree/main/train |
| 英文预训练 | Pile 22个高质量数据集混合的预训练数据集800G,全量开放下载 | https://pile.eleuther.ai/ |
| 通用预训练 | UER整理CLUECorpusSmall+News Commentary中英 | https://github.com/dbiir/UER-py/wiki/%E9%A2%84%E8%AE%AD%E7%BB%83%E6%95%B0%E6%8D%AE |
| 中文预训练 | 智源人工智能开源的wudao 200G预训练数据 | https://github.com/BAAI-WuDao/WuDaoMM |
| 中文预训练 | 里屋社区发起开源力量收集中文互联网语料集MNBVC目标是对标ChatGPT的40T | https://github.com/esbatmop/MNBVC |
| 中文预训练 | 复旦开源15万中文图书下载和抽取方案 | https://github.com/FudanNLPLAB/CBook-150K |
| 中文预训练 | 书生万卷数据集来自公开网页多模态数据集，包括文本，图文和视频，其中文本1T，图文150G | https://opendatalab.org.cn/OpenDataLab/WanJuan1_dot_0 |
| 中文预训练 | 昆仑天工开源3.2TB中英语料 | https://github.com/SkyworkAI/Skywork |
| 中文预训练 | 浪潮开源的用于Yuan1.0训练的预训练中文语料 | https://www.airyuan.cn/home |
| 领域预训练 | 度小满开源60G金融预训练语料 | https://github.com/Duxiaoman-DI/XuanYuan |
| 领域预训练 | 首个中文科学文献数据集CSL,也有多种NLP任务数据 | https://github.com/ydli-ai/CSL |
| 平行语料 | news-commentary中英平行语料，用于中英间知识迁移 | https://data.statmt.org/news-commentary/v15/training/ |
| 多源数据集整合 | opendatalab整合了预训练阶段的多个数据源 | https://opendatalab.org.cn/?industry=9821&source=JUU3JTlGJUE1JUU0JUI5JThF |
| Tool-搜索增强 | webCPM开源的和搜索工具进行交互问答的数据集，包括网页抽取式摘要，多事实内容回答等人工标注数据 | https://github.com/thunlp/WebCPM |
| Tool-多工具 | BmTools开源的多工具调用指令数据集 | https://github.com/OpenBMB/BMTools |
| Tool-多工具 | AgentInstruct包含6项Agent任务，包括REACT式COT标注 | https://thudm.github.io/AgentTuning/ |
| Tool-多工具 | MSAgent-Bench 大模型调用数据集 598k训练数据 | https://modelscope.cn/datasets/damo/MSAgent-Bench/summary |
| Tool-多工具 | MOSS开源的知识搜索，文生图，计算器，解方程等4个插件的30万条多轮对话数据 | https://github.com/OpenLMLab/MOSS#%E6%95%B0%E6%8D%AE |
| NL2SQL | DB-GPT-Hub梳理了多源text-to-sql数据集 | https://github.com/eosphoros-ai/DB-GPT-Hub |
| 长文本 | 清华开源的长文本对齐数据集LongAlign-10k | https://huggingface.co/datasets/THUDM/LongAlign-10k |
AIGC
搜索
通用搜索
- 秘塔搜索: 融合了脑图，表格多模态问答的搜索应用
- You.COM : 支持多种检索增强问答模式
- Walles.AI: 融合了图像聊天，文本聊天，chatpdf，web-copilot等多种功能的智能助手
- webpilot.ai 比ChatGPT 自带的 Web Browsing更好用的浏览器检索插件，更适用于复杂搜索场景，也开通api调用了
- New Bing：需要科学上网哦
- Perplexity.ai: 同样需要科学上网，感觉比Bing做的更好的接入ChatGPT的神奇搜索引擎，在Bing之外还加入了相关推荐和追问
- sider.ai: 支持多模型浏览器插件对话和多模态交互操作
代码搜索
- devv.ai: 基于微调llama2 + RAG搭建的属于程序员的搜索引擎
- Phind: 面向开发人员的AI搜索引擎
知识管理
- glean: 企业知识搜索和项目管理类的搜索初创公司，帮助员工快速定位信息，帮助公司整合信息
- Mem: 个人知识管理，例如知识图谱，已获openai融资
ChatDoc
- Kimi-Chat: 长长长长文档理解无敌的Kimi-Chat，单文档总结多文档结构化对比，无所不能，多长都行！
- ChatDoc:ChatPDF升级版，需要科学上网，增加了表格类解析，支持选择区域的问答，在PDF识别上做的很厉害
- AskyourPdf: 同样是上传pdf进行问答和摘要的应用
- DocsGPT: 比较早出来的Chat DOC通用方案
- ChatPDF: 国内的ChatPDF, 上传pdf后，会给出文章的Top5可能问题，然后对话式从文档中进行问答和检索，10s读3万字
- AlphaBox: 从个人文件夹管理出发的文档问答工具
论文研究: 日度更新，观点总结，
- SCISPACE: 论文研究的白月光，融合了全库搜索问答，以及个人上传PDF构建知识库问答。同样支持相关论文发现，和论文划词解读。并且解读内容可以保存到notebook中方便后续查找，可以说是产品和算法强强联合了。
- Consensus: AI加持的论文搜素，多论文总结，观点对比工具。产品排名巨高，但个人感觉搜索做的有提升空间
- Aminer: 论文搜索，摘要，问答，搜索关键词符号化改写；但论文知识库问答有些幻觉严重
- cool.paper: 苏神开发的基于kimi的论文阅读网站
- OpenRead: 国内产品，面向论文写作，阅读场景，可以帮助生成文献综述，以及提供和NotionAI相似的智能Markdown用于写作
- ChatPaper: 根据输入关键词，自动在arxiv上下载最新的论文，并对论文进行摘要总结，可以在huggingface上试用
- researchgpt: 和ChatPDF类似，支持arivx论文下载，加载后对话式获取论文重点
- ChatGPT-academic: 又是一个基于gradio实现的paper润色，摘要等功能打包的实现,不少功能可以借鉴
- BriefGPT: 日更Arxiv论文，并对论文进行摘要，关键词抽取，帮助研究者了解最新动态, UI不错哟
写作效率工具类
- 赛博马良:题如其名，可定制AI员工24小时全网抓取关注的创作选题，推送给小编进行二次创作
- Miracleplus: 全AI Agent负责运营的Hacker News网站
- ChatMind: chatgpt生成思维导图，模板很丰富，泛化性也不错，已经被XMind收购了
- 范文喵写作: 范文喵写作工具，选题，大纲，写作全流程
- WriteSonic：AI写作，支持对话和定向创作如广告文案，商品描述, 支持Web检索是亮点，支持中文
- copy.ai: WriteSonic竞品，亮点是像论文引用一样每句话都有对应网站链接，可以一键复制到右边的创作Markdown，超级好用！
- NotionAI：智能Markdown，适用真相！在创作中用command调用AI辅助润色，扩写，检索内容，给创意idea
- Hix-AI: 同时提供copilot模式和综合写作模式
- AI-Write： 个人使用感较好的流程化写作工具
- Jasper: 同上，全是竞品哈哈
- copy.down: 中文的营销文案生成，只能定向创作，支持关键词到文案的生成
- Weaver AI: 波形智能开发的内容创作app，支持多场景写作
- ChatExcel: 指令控制excel计算，对熟悉excel的有些鸡肋，对不熟悉的有点用
- mindShow：免费+付费的PPT制作工具，自定义PPT模板还不够好
金融垂直领域
- Alpha: ChatGPT加持的金融app，支持个股信息查询，资产分析诊断，财报汇总etc
- Composer：量化策略和AI的结合，聊天式+拖拽式投资组合构建和回测
- Finalle.ai: 实时金融数据流接入大模型
- ScopeChat:虚拟币应用，整个对话类似ChatLaw把工具组件嵌入了对话中
- AInvest：个股投资，融合BI分析，广场讨论区（有演变成雪球热度指数的赶脚)
- Reportify: 金融领域公司公告，新闻，电话会的问答和摘要总结
私人助理&聊天
- Mr.-Ranedeer-: 基于prompt和GPT-4的强大能力提供个性化学习环境，个性化出题+模型解答
- AI Topiah: 聆心智能AI角色聊天，和路飞唠了两句，多少有点中二之魂在燃烧
- chatbase: 情感角色聊天，还没尝试
- Vana: virtual DNA, 通过聊天创建虚拟自己！概念很炫
Agent
- NexusGPT: AutoGPT可以出来工作了，第一个全AI Freelance平台
- cognosys: 全网最火的web端AutoGPT，不过咋说呢试用了下感觉下巴要笑掉了，不剧透去试试你就知道
- godmode：可以进行人为每一步交互的的AutoGPT
- agentgpt: 基础版AutoGPT
视频拆条总结
- Eightify: chrome插件，节省观看长视频的时间，立即获取关键思想，分模块总结+时间戳摘要
- BibiGPT: Bilibli视频内容一键总结，多模态文档
代码copilot & BI工具
- AutoDev: AI编程辅助工具
- AlphaCodium: Flow Engineering提高代码整体通过率
- Codium: 开源的编程Copilot来啦
- Copilot: 要付费哟
- Fauxpilot: copilot本地开源替代
- Codeium: Copilot替代品，有免费版本支持各种plugin !
- ai2sql: text2sql老牌公司，相比sqltranslate功能更全面，支持SQL 语法检查、格式化和生成公式
- chat2query: text2sql 相比以上两位支持更自然的文本指令，以及更复杂的数据分析类的sql生成
- OuterBase: text2sql 设计风格很吸睛！电子表格结合mysql和dashboard，更适合数据分析宝宝
- Chat2DB:智能的通用数据库SQL客户端和报表工具
- ChatBI:网易数帆发布ChatBI对话数据分析平台
- Kyligence Copilot:Kyligence发布一站式指标平台的 AI 数智助理,支持对话式指标搜索，异动归因等等
- Wolverine: 代码自我debug的python脚本
多模态生成
- dreamstudio.ai: 开创者，Stable Difussion， 有试用quota
- midjourney: 开创者，艺术风格为主
- Dall.E: 三巨头这就凑齐了
- ControlNet: 为绘画创作加持可控性
- gemo.ai: 多模态聊天机器人，包括文本，图像，视频生成
- storybird: 根据提示词生成故事绘本，还可以售卖
- Magnific.ai: 两个人的团队做出的AI图片精修师
- Morph Studio: Stability AI入场视频制作
- Gamma: PPT制作神器，ProductHunt月度排名Number1
Resources
GPTs应用导航
- SimilarGPTs：全网AI产品流量大全：
- GPTSeek: 大家投票得出的最有价值的GPT应用
- ProductHunt: 技术产品网站，各类热门AI技术产品的集散地
- AI-Product-Index
- AI-Products-All-In-One
- TheRunDown: GPT应用分类
- Awesome AI Agents：Agent应用收藏
- GPT Demo
- AI-Bot各类工具导航
Prompt和其他教程类
- OpenAI Cookbook: 提供OpenAI模型使用示例 ⭐️
- PromptPerfect:用魔法打败魔法，输入原始提示词，模型进行定向优化，试用后我有点沉默了，可以定向支持不同使用prompt的模型如Difussion，ChatGPT， Dalle等
- ClickPrompt: 为各种prompt加持的工具生成指令包括Difussion，chatgptdeng, 需要OpenAI Key
- ChatGPT ShortCut：提供各式场景下的Prompt范例，范例很全，使用后可以点赞！ ⭐️
- Full ChatGPT Prompts + Resources: 各种尝尽的prompt范例，和以上场景有所不同
- learning Prompt: prompt engineering超全教程，和落地应用收藏，包括很多LLM调用Agent的高级场景 ⭐️
- The art of asking chatgpt for high quality answers: 如何写Prompt指令出书了，链接是中文翻译的版本，比较偏基础使用
- Prompt-Engineer-Guide: 同learnig prompt类的集成教程，互相引用可还行？！分类索引做的更好些 ⭐️
- AI Alignment Forum: RLHF等对齐相关最新论文和观点的讨论论坛
- Langchain: Chat with your data:吴恩达LLM实践课程
- 构筑大语言模型应用：应用开发与架构设计: 一本关于 LLM 在真实世界应用的开源电子书
- Large Language Models: Application through Production: 大模型应用Edx出品的课程
- Minbpe: Karpathy大佬离职openai后整了个分词器的教学代码
- LLM-VIZ: 大模型结构可视化支持GPT系列
书籍和博客类
- OpenAI ChatGPT Intro
- OpenAI InstructGPT intro
- AllenAI ChatGPT能力解读：How does GPT Obtain its Ability? Tracing Emergent Abilities of Language Models to their Sources ⭐️
- Huggingface ChatGPT能力解读：The techniques behind ChatGPT: RLHF, IFT, CoT, Red teaming, and more
- Stephen Wolfram ChatGPT能力解读: What Is ChatGPT Doing and Why Does It Work?
- Chatgpt相关解读汇总
- AGI历史与现状
- 张俊林 通向AGI之路：大型语言模型（LLM）技术精要
- 知乎回答 OpenAI 发布 GPT-4，有哪些技术上的优化或突破?
- 追赶ChatGPT的难点与平替
- 压缩即泛化，泛化即智能 ⭐️
- LLM Powered Autonomous Agents ⭐️
- All You Need to Know to Build Your First LLM App ⭐️
- GPT-4 Architecture, Infrastructure, Training Dataset, Costs, Vision, MoE
- OpenAI研究员出书：为什么伟大不能被计划:
- 拾象投研机构对LLM的调研报告（文中有两次PPT的申请链接）:
- 启明创投State of Generative AI 2023
- How to Use AI to Do Stuff: An Opinionated Guide
- Llama 2: an incredible open LLM
- Wolfram语言之父新书：这就是ChatGPT
- 谷歌出品：对大模型领悟能力的一些探索很有意思
Do Machine Learning Models Memorize or Generalize? ⭐️ - 符尧大佬系列新作 An Initial Exploration of Theoretical Support for Language Model Data Engineering. Part 1: Pretraining
- 奇绩创坛2023秋季路演日上创新LLM项目一览
- The Power of Prompting微软首席科学家对prompt在垂直领域使用的观点
- The Bitter Lesson 强化学习之父总结的AI研究的经验教训
会议&访谈类
- 麻省理工科技采访OpenAI工程师
- 陆奇最新演讲实录：我的大模型世界观｜第十四期
- OpenAI首席科学家最新讲座解读LM无监督预训练学了啥 An observation on Generalization ⭐️
- The Complete Beginners Guide To Autonomous Agents: Octane AI创始人 Matt Schlicht发表的关于人工智能代理的一些思考
- Large Language Models (in 2023) OpenAI科学家最新大模型演讲
- OpenAI闭门会议DevDay视频 - A survey of Techniques for Maximizing LLM performance，无法可搜标题找笔记
- 月之暗面杨植麟专访,值得细读
一分钟上手系列：https://blog.csdn.net/u014374009/category_12451843.html
Papers
paper List
- https://github.com/dongguanting/In-Context-Learning_PaperList
- https://github.com/thunlp/PromptPapers
- https://github.com/Timothyxxx/Chain-of-ThoughtsPapers
- https://github.com/thunlp/ToolLearningPapers
- https://github.com/MLGroupJLU/LLM-eval-survey
- https://github.com/thu-coai/PaperForONLG
综述
- A Survey of Large Language Models
- Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing ⭐️
- Paradigm Shift in Natural Language Processing
- Pre-Trained Models: Past, Present and Future
- What Language Model Architecture and Pretraining objects work best for zero shot generalization ⭐️
- Towards Reasoning in Large Language Models: A Survey
- Reasoning with Language Model Prompting: A Survey ⭐️
- An Overview on Language Models: Recent Developments and Outlook ⭐️
- A Survey of Large Language Models[6.29更新版]
- Unifying Large Language Models and Knowledge Graphs: A Roadmap
- Augmented Language Models: a Survey ⭐️
- Domain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey
- Challenges and Applications of Large Language Models
- The Rise and Potential of Large Language Model Based Agents: A Survey
- Large Language Models for Information Retrieval: A Survey
- AI Alignment: A Comprehensive Survey
- Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomy of Methods, Benchmarks, and Applications
- Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook
- A Survey on Language Models for Code
- Model-as-a-Service (MaaS): A Survey
大模型能力探究
- In Context Learning
- LARGER LANGUAGE MODELS DO IN-CONTEXT LEARNING DIFFERENTLY
- How does in-context learning work? A framework for understanding the differences from traditional supervised learning
- Why can GPT learn in-context? Language Model Secretly Perform Gradient Descent as Meta-Optimizers ⭐️
- Rethinking the Role of Demonstrations What Makes incontext learning work? ⭐️
- Trained Transformers Learn Linear Models In-Context
- In-Context Learning Creates Task Vectors
- 涌现能力
- Sparks of Artificial General Intelligence: Early experiments with GPT-4
- Emerging Ability of Large Language Models ⭐️
- LANGUAGE MODELS REPRESENT SPACE AND TIME
- Are Emergent Abilities of Large Language Models a Mirage?
- 能力评估
- IS CHATGPT A GENERAL-PURPOSE NATURAL LANGUAGE PROCESSING TASK SOLVER?
- Can Large Language Models Infer Causation from Correlation?
- Holistic Evaluation of Language Model
- Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond
- Theory of Mind May Have Spontaneously Emerged in Large Language Models
- Beyond The Imitation Game: Quantifying And Extrapolating The Capabilities Of Language Models
- Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations
- Demystifying GPT Self-Repair for Code Generation
- Evidence of Meaning in Language Models Trained on Programs
- Can Explanations Be Useful for Calibrating Black Box Models
- On the Robustness of ChatGPT: An Adversarial and Out-of-distribution Perspective
- Language acquisition: do children and language models follow similar learning stages?
- 领域能力
- Capabilities of GPT-4 on Medical Challenge Problems
- Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine
Prompt Tunning范式
- Tunning Free Prompt
- GPT2: Language Models are Unsupervised Multitask Learners
- GPT3: Language Models are Few-Shot Learners ⭐️
- LAMA: Language Models as Knowledge Bases?
- AutoPrompt: Eliciting Knowledge from Language Models
- Fix-Prompt LM Tunning
- T5: Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
- PET-TC(a): Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference ⭐️
- PET-TC(b): PETSGLUE It’s Not Just Size That Matters Small Language Models are also few-shot learners
- GenPET: Few-Shot Text Generation with Natural Language Instructions
- LM-BFF: Making Pre-trained Language Models Better Few-shot Learners ⭐️
- ADEPT: Improving and Simplifying Pattern Exploiting Training
- Fix-LM Prompt Tunning
- Prefix-tuning: Optimizing continuous prompts for generation
- Prompt-tunning: The power of scale for parameter-efficient prompt tuning ⭐️
- P-tunning: GPT Understands Too ⭐️
- WARP: Word-level Adversarial ReProgramming
- LM + Prompt Tunning
- P-tunning v2: Prompt Tuning Can Be Comparable to Fine-tunning Universally Across Scales and Tasks
- PTR: Prompt Tuning with Rules for Text Classification
- PADA: Example-based Prompt Learning for on-the-fly Adaptation to Unseen Domains
- Fix-LM Adapter Tunning
- LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS ⭐️
- LST: Ladder Side-Tuning for Parameter and Memory Efficient Transfer Learning
- Parameter-Efficient Transfer Learning for NLP
- INTRINSIC DIMENSIONALITY EXPLAINS THE EFFECTIVENESS OF LANGUAGE MODEL FINE-TUNING
主流LLMS和预训练
- GLM-130B: AN OPEN BILINGUAL PRE-TRAINED MODEL
- PaLM: Scaling Language Modeling with Pathways
- PaLM 2 Technical Report
- GPT-4 Technical Report
- Backpack Language Models
- LLaMA: Open and Efficient Foundation Language Models
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning
- OpenBA: An Open-sourced 15B Bilingual Asymmetric seq2seq Model Pre-trained from Scratch
- Mistral 7B
- Ziya2: Data-centric Learning is All LLMs Need
- MEGABLOCKS: EFFICIENT SPARSE TRAINING WITH MIXTURE-OF-EXPERTS
- TUTEL: ADAPTIVE MIXTURE-OF-EXPERTS AT SCALE
- Phi1- Textbooks Are All You Need ⭐️
- Phi1.5- Textbooks Are All You Need II: phi-1.5 technical report
- Gemini: A Family of Highly Capable Multimodal Models
- In-Context Pretraining: Language Modeling Beyond Document Boundaries
- LLAMA PRO: Progressive LLaMA with Block Expansion
- QWEN TECHNICAL REPORT
指令微调&对齐 (instruction_tunning)
- 经典方案
- Flan: FINETUNED LANGUAGE MODELS ARE ZERO-SHOT LEARNERS ⭐️
- Flan-T5: Scaling Instruction-Finetuned Language Models
- ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning
- Instruct-GPT: Training language models to follow instructions with human feedback ⭐️
- T0: MULTITASK PROMPTED TRAINING ENABLES ZERO-SHOT TASK GENERALIZATION
- Natural Instructions: Cross-Task Generalization via Natural Language Crowdsourcing Instructions
- Tk-INSTRUCT: SUPER-NATURALINSTRUCTIONS: Generalization via Declarative Instructions on 1600+ NLP Tasks
- ZeroPrompt: Scaling Prompt-Based Pretraining to 1,000 Tasks Improves Zero-shot Generalization
- Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor
- INSTRUCTEVAL Towards Holistic Evaluation of Instrucion-Tuned Large Language Models
- 更少，质量更高、更多样的指令数据带来质变
- LIMA: Less Is More for Alignment ⭐️
- Maybe Only 0.5% Data is Needed: A Preliminary Exploration of Low Training Data Instruction Tuning
- AlpaGasus: Training A Better Alpaca with Fewer Data
- InstructionGPT-4: A 200-Instruction Paradigm for Fine-Tuning MiniGPT-4
- Instruction Mining: High-Quality Instruction Data Selection for Large Language Models
- Visual Instruction Tuning with Polite Flamingo
- Exploring the Impact of Instruction Data Scaling on Large Language Models: An Empirical Study on Real-World Use Cases
- Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
- 新对齐/微调方案
- WizardLM: Empowering Large Language Models to Follow Complex Instructions
- Becoming self-instruct: introducing early stopping criteria for minimal instruct tuning
- Self-Alignment with Instruction Backtranslation ⭐️
- Mixture-of-Experts Meets Instruction Tuning:A Winning Combination for Large Language Models
- Goat: Fine-tuned LLaMA Outperforms GPT-4 on Arithmetic Tasks
- PROMPT2MODEL: Generating Deployable Models from Natural Language Instructions
- OpinionGPT: Modelling Explicit Biases in Instruction-Tuned LLMs
- Improving Language Model Negotiation with Self-Play and In-Context Learning from AI Feedback
- Human-like systematic generalization through a meta-learning neural network
- Magicoder: Source Code Is All You Need
- Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models
- Generative Representational Instruction Tuning
- 指令数据生成
- APE: LARGE LANGUAGE MODELS ARE HUMAN-LEVEL PROMPT ENGINEERS ⭐️
- SELF-INSTRUCT: Aligning Language Model with Self Generated Instructions ⭐️
- iPrompt: Explaining Data Patterns in Natural Language via Interpretable Autoprompting
- Flipped Learning: Guess the Instruction! Flipped Learning Makes Language Models Stronger Zero-Shot Learners
- Fairness-guided Few-shot Prompting for Large Language Models
- Instruction induction: From few examples to natural language task descriptions .
- SELF-QA Unsupervised Knowledge Guided alignment.
- GPT Self-Supervision for a Better Data Annotator
- The Flan Collection Designing Data and Methods
- Self-Consuming Generative Models Go MAD
- InstructEval: Systematic Evaluation of Instruction Selection Methods
- Overwriting Pretrained Bias with Finetuning Data
- Improving Text Embeddings with Large Language Models
- 如何降低通用能力损失
- How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition
- TWO-STAGE LLM FINE-TUNING WITH LESS SPECIALIZATION AND MORE GENERALIZATION
- 微调经验/实验报告
- BELLE: Exploring the Impact of Instruction Data Scaling on Large Language Models: An Empirical Study on Real-World Use Cases
- Baize: Baize: An Open-Source Chat Model with Parameter-Efficient Tuning on Self-Chat Data
- A Comparative Study between Full-Parameter and LoRA-based Fine-Tuning on Chinese Instruction Data for Large LM
- Exploring ChatGPT’s Ability to Rank Content: A Preliminary Study on Consistency with Human Preferences
- Towards Better Instruction Following Language Models for Chinese: Investigating the Impact of Training Data and Evaluation
- Others
- Crosslingual Generalization through Multitask Finetuning
- Cross-Task Generalization via Natural Language Crowdsourcing Instructions
- UNIFIEDSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
- PromptSource: An Integrated Development Environment and Repository for Natural Language Prompts
- ROLELLM: BENCHMARKING, ELICITING, AND ENHANCING ROLE-PLAYING ABILITIES OF LARGE LANGUAGE MODELS
对话模型
- LaMDA: Language Models for Dialog Applications
- Sparrow: Improving alignment of dialogue agents via targeted human judgements ⭐️
- BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage
- How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
- DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI
- Enhancing Chat Language Models by Scaling High-quality Instructional Conversations
- DiagGPT: An LLM-based Chatbot with Automatic Topic Management for Task-Oriented Dialogue
思维链 (prompt_chain_of_thought)
- 基础&进阶用法
- [zero-shot-COT] Large Language Models are Zero-Shot Reasoners ⭐️
- [few-shot COT] Chain of Thought Prompting Elicits Reasoning in Large Language Models ⭐️
- SELF-CONSISTENCY IMPROVES CHAIN OF THOUGHT REASONING IN LANGUAGE MODELS
- LEAST-TO-MOST PROMPTING ENABLES COMPLEX REASONING IN LARGE LANGUAGE MODELS ⭐️
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
- Decomposed Prompting A MODULAR APPROACH FOR Solving Complex Tasks
- Successive Prompting for Decomposing Complex Questions
- Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework
- Beyond Chain-of-Thought, Effective Graph-of-Thought Reasoning in Large Language Models
- Tree-of-Mixed-Thought: Combining Fast and Slow Thinking for Multi-hop Visual Reasoning
- LAMBADA: Backward Chaining for Automated Reasoning in Natural Language
- Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
- Graph of Thoughts: Solving Elaborate Problems with Large Language Models
- Progressive-Hint Prompting Improves Reasoning in Large Language Models
- LARGE LANGUAGE MODELS CAN LEARN RULES
- DIVERSITY OF THOUGHT IMPROVES REASONING ABILITIES OF LARGE LANGUAGE MODELS
- From Complex to Simple: Unraveling the Cognitive Tree for Reasoning with Small Language Models
- Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models
- LARGE LANGUAGE MODELS AS OPTIMIZERS
- 分领域COT [Math, Code, Tabular, QA]
- Solving Quantitative Reasoning Problems with Language Models
- SHOW YOUR WORK: SCRATCHPADS FOR INTERMEDIATE COMPUTATION WITH LANGUAGE MODELS
- Solving math word problems with processand outcome-based feedback
- CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning
- T-SciQ: Teaching Multimodal Chain-of-Thought Reasoning via Large Language Model Signals for Science Question Answering
- LEARNING PERFORMANCE-IMPROVING CODE EDITS
- Large Language Models are Versatile Decomposers: Decompose Evidence and Questions for Table-based Reasoning
- Tab-CoT: Zero-shot Tabular Chain of Thought
- Chain of Code: Reasoning with a Language Model-Augmented Code Emulator
- 原理分析
- Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters ⭐️
- TEXT AND PATTERNS: FOR EFFECTIVE CHAIN OF THOUGHT IT TAKES TWO TO TANGO
- Towards Revealing the Mystery behind Chain of Thought: a Theoretical Perspective
- Large Language Models Can Be Easily Distracted by Irrelevant Context
- Chain-of-Thought Reasoning Without Prompting
- 小模型COT蒸馏
- Specializing Smaller Language Models towards Multi-Step Reasoning ⭐️
- Teaching Small Language Models to Reason
- Large Language Models are Reasoning Teachers
- Distilling Reasoning Capabilities into Smaller Language Models
- The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuning
- COT样本自动构建/选择
- STaR: Self-Taught Reasoner Bootstrapping ReasoningWith Reasoning
- AutoCOT：AUTOMATIC CHAIN OF THOUGHT PROMPTING IN LARGE LANGUAGE MODELS
- Large Language Models Can Self-Improve
- Active Prompting with Chain-of-Thought for Large Language Models
- COMPLEXITY-BASED PROMPTING FOR MULTI-STEP REASONING
- others
- OlaGPT Empowering LLMs With Human-like Problem-Solving abilities
- Challenging BIG-Bench tasks and whether chain-of-thought can solve them
- Large Language Models are Better Reasoners with Self-Verification
- ThoughtSource A central hub for large language model reasoning data
- Two Failures of Self-Consistency in the Multi-Step Reasoning of LLMs
RLHF
- Deepmind
- Teaching language models to support answers with verified quotes
- sparrow, Improving alignment of dialogue agents via targetd human judgements ⭐️
- STATISTICAL REJECTION SAMPLING IMPROVES PREFERENCE OPTIMIZATION
- Reinforced Self-Training (ReST) for Language Modeling
- SLiC-HF: Sequence Likelihood Calibration with Human Feedback
- CALIBRATING SEQUENCE LIKELIHOOD IMPROVES CONDITIONAL LANGUAGE GENERATION
- REWARD DESIGN WITH LANGUAGE MODELS
- Final-Answer RL Solving math word problems with processand outcome-based feedback
- Solving math word problems with process- and outcome-based feedback
- openai
- PPO: Proximal Policy Optimization Algorithms ⭐️
- Deep Reinforcement Learning for Human Preference
- Fine-Tuning Language Models from Human Preferences
- learning to summarize from human feedback
- InstructGPT: Training language models to follow instructions with human feedback ⭐️
- Scaling Laws for Reward Model Over optimization ⭐️
- WEAK-TO-STRONG GENERALIZATION: ELICITING STRONG CAPABILITIES WITH WEAK SUPERVISION ⭐️
- PRM：Let’s verify step by step
-
- OpenAI Super Alignment Blog
- Anthropic
- A General Language Assistant as a Laboratory for Alignmen
- Red Teaming Language Models to Reduce Harms Methods,Scaling Behaviors and Lessons Learned
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback ⭐️
- Constitutional AI Harmlessness from AI Feedback ⭐️
- Pretraining Language Models with Human Preferences
- The Capacity for Moral Self-Correction in Large Language Models
- Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainin
- AllenAI, RL4LM：IS REINFORCEMENT LEARNING (NOT) FOR NATURAL LANGUAGE PROCESSING BENCHMARKS
- 改良方案
- RRHF: Rank Responses to Align Language Models with Human Feedback without tears
- Chain of Hindsight Aligns Language Models with Feedback
- AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
- RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment
- RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback
- Training Socially Aligned Language Models in Simulated Human Society
- RAIN: Your Language Models Can Align Themselves without Finetuning
- Generative Judge for Evaluating Alignment
- PEERING THROUGH PREFERENCES: UNRAVELING FEEDBACK ACQUISITION FOR ALIGNING LARGE LANGUAGE MODELS
- SALMON: SELF-ALIGNMENT WITH PRINCIPLE-FOLLOWING REWARD MODELS
- Large Language Model Unlearning ⭐️
- ADVERSARIAL PREFERENCE OPTIMIZATION ⭐️
- Preference Ranking Optimization for Human Alignment
- A Long Way to Go: Investigating Length Correlations in RLHF
- ENABLE LANGUAGE MODELS TO IMPLICITLY LEARN SELF-IMPROVEMENT FROM DATA
- REWARD MODEL ENSEMBLES HELP MITIGATE OVEROPTIMIZATION
- LEARNING OPTIMAL ADVANTAGE FROM PREFERENCES AND MISTAKING IT FOR REWARD
- ULTRAFEEDBACK: BOOSTING LANGUAGE MODELS WITH HIGH-QUALITY FEEDBACK
- MOTIF: INTRINSIC MOTIVATION FROM ARTIFICIAL INTELLIGENCE FEEDBACK
- STABILIZING RLHF THROUGH ADVANTAGE MODEL AND SELECTIVE REHEARSAL
- Shepherd: A Critic for Language Model Generation
- LEARNING TO GENERATE BETTER THAN YOUR LLM
- Fine-Grained Human Feedback Gives Better Rewards for Language Model Training
- Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- HIR The Wisdom of Hindsight Makes Language Models Better Instruction Followers
- RL探究
- UNDERSTANDING THE EFFECTS OF RLHF ON LLM GENERALISATION AND DIVERSITY
- A LONG WAY TO GO: INVESTIGATING LENGTH CORRELATIONS IN RLHF
- THE TRICKLE-DOWN IMPACT OF REWARD (IN-)CONSISTENCY ON RLHF
- Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
- HUMAN FEEDBACK IS NOT GOLD STANDARD
- CONTRASTIVE POST-TRAINING LARGE LANGUAGE MODELS ON DATA CURRICULUM
LLM Agent 让模型使用工具 (llm_agent)
- A Survey on Large Language Model based Autonomous Agents
- PERSONAL LLM AGENTS: INSIGHTS AND SURVEY ABOUT THE CAPABILITY, EFFICIENCY AND SECURITY
- 基于prompt通用方案
- ReAct: SYNERGIZING REASONING AND ACTING IN LANGUAGE MODELS ⭐️
- Self-ask: MEASURING AND NARROWING THE COMPOSITIONALITY GAP IN LANGUAGE MODELS ⭐️
- MRKL SystemsA modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning
- PAL: Program-aided Language Models
- ART: Automatic multi-step reasoning and tool-use for large language models
- ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models ⭐️
- Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions
- Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models ⭐️
- Faithful Chain-of-Thought Reasoning
- Reflexion: Language Agents with Verbal Reinforcement Learning ⭐️
- Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework
- RestGPT: Connecting Large Language Models with Real-World RESTful APIs
- ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models
- InstructTODS: Large Language Models for End-to-End Task-Oriented Dialogue Systems
- TPTU: Task Planning and Tool Usage of Large Language Model-based AI Agents
- ControlLLM: Augment Language Models with Tools by Searching on Graphs
- Reflexion: an autonomous agent with dynamic memory and self-reflection
- AutoAgents: A Framework for Automatic Agent Generation
- GitAgent: Facilitating Autonomous Agent with GitHub by Tool Extension
- PreAct: Predicting Future in ReAct Enhances Agent’s Planning Ability
- TOOLLLM: FACILITATING LARGE LANGUAGE MODELS TO MASTER 16000+ REAL-WORLD APIS
- 基于微调通用方案
- TALM: Tool Augmented Language Models
- Toolformer: Language Models Can Teach Themselves to Use Tools ⭐️
- Tool Learning with Foundation Models
- Tool Maker：Large Language Models as Tool Maker
- TaskMatrix.AI: Completing Tasks by Connecting Foundation Models with Millions of APIs
- AgentTuning: Enabling Generalized Agent Abilities for LLMs
- SWIFTSAGE: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks
- FireAct: Toward Language Agent Fine-tuning
- Pangu-Agent: A Fine-Tunable Generalist Agent with Structured Reasoning
- REST MEETS REACT: SELF-IMPROVEMENT FOR MULTI-STEP REASONING LLM AGENT
- Efficient Tool Use with Chain-of-Abstraction Reasoning
- 调用模型方案
- HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in HuggingFace
- Gorilla：Large Language Model Connected with Massive APIs ⭐️
- OpenAGI: When LLM Meets Domain Experts
- 垂直领域
- WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents
- ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings
- ChemCrow Augmenting large language models with chemistry tools
- Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow
- Demonstration of InsightPilot: An LLM-Empowered Automated Data Exploration System
- GeneGPT: Augmenting Large Language Models with Domain Tools for Improved Access to Biomedical Information
- PointLLM: Empowering Large Language Models to Understand Point Clouds
- Interpretable Long-Form Legal Question Answering with Retrieval-Augmented Large Language Models
- Generating Explanations in Medical Question-Answering by Expectation Maximization Inference over Evidence
- CarExpert: Leveraging Large Language Models for In-Car Conversational Question Answering
- A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist
- 评估
- Evaluating Verifiability in Generative Search Engines
- Mind2Web: Towards a Generalist Agent for the Web
- Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions
- API-Bank: A Benchmark for Tool-Augmented LLMs
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- Automatic Evaluation of Attribution by Large Language Models
- Benchmarking Large Language Models in Retrieval-Augmented Generation
- ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems
- MultiAgent
- Generative Agents: Interactive Simulacra of Human Behavior ⭐️
- AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors in Agents
- CAMEL: Communicative Agents for “Mind” Exploration of Large Scale Language Model Society ⭐️
- Exploring Large Language Models for Communication Games: An Empirical Study on Werewolf
- Communicative Agents for Software Development ⭐️
- METAAGENTS: SIMULATING INTERACTIONS OF HUMAN BEHAVIORS FOR LLM-BASED TASK-ORIENTED COORDINATION VIA COLLABORATIVE GENERATIVE AGENTS
- LET MODELS SPEAK CIPHERS: MULTIAGENT DEBATE THROUGH EMBEDDINGS
- MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning
- War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars
- More Agents Is All You Need
- 自主学习和探索
- AppAgent: Multimodal Agents as Smartphone Users
- Investigate-Consolidate-Exploit: A General Strategy for Inter-Task Agent Self-Evolution
- LLMs in the Imaginarium: Tool Learning through Simulated Trial and Error
- 其他
- LLM+P: Empowering Large Language Models with Optimal Planning Proficiency
- Inference with Reference: Lossless Acceleration of Large Language Models
- RecallM: An Architecture for Temporal Context Understanding and Question Answering
- LLaMA Rider: Spurring Large Language Models to Explore the Open World
RAG
- WebGPT：Browser-assisted question-answering with human feedback
- WebGLM: Towards An Efficient Web-Enhanced Question Answering System with Human Preferences
- WebCPM: Interactive Web Search for Chinese Long-form Question Answering ⭐️
- REPLUG: Retrieval-Augmented Black-Box Language Models ⭐️
- Query Rewriting for Retrieval-Augmented Large Language Models
- RETA-LLM: A Retrieval-Augmented Large Language Model Toolkit
- Atlas: Few-shot Learning with Retrieval Augmented Language Models
- RRAML: Reinforced Retrieval Augmented Machine Learning
- Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation
- PDFTriage: Question Answering over Long, Structured Documents
- SELF-RAG: LEARNING TO RETRIEVE, GENERATE, AND CRITIQUE THROUGH SELF-REFLECTION ⭐️
- Walking Down the Memory Maze: Beyond Context Limit through Interactive Reading ⭐️
- Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP
- Search-in-the-Chain: Towards Accurate, Credible and Traceable Large Language Models for Knowledge-intensive Tasks
- Active Retrieval Augmented Generation
- kNN-LM Does Not Improve Open-ended Text Generation
- Can Retriever-Augmented Language Models Reason? The Blame Game Between the Retriever and the Language Model
- Query2doc: Query Expansion with Large Language Models ⭐️
- RLCF：Aligning the Capabilities of Large Language Models with the Context of Information Retrieval via Contrastive Feedback
- Augmented Embeddings for Custom Retrievals
- DORIS-MAE: Scientific Document Retrieval using Multi-level Aspect-based Queries
- Learning to Filter Context for Retrieval-Augmented Generation
- THINK-ON-GRAPH: DEEP AND RESPONSIBLE REASON- ING OF LARGE LANGUAGE MODEL ON KNOWLEDGE GRAPH
- RA-DIT: RETRIEVAL-AUGMENTED DUAL INSTRUCTION TUNING
- Query Expansion by Prompting Large Language Models ⭐️
- CHAIN-OF-NOTE: ENHANCING ROBUSTNESS IN RETRIEVAL-AUGMENTED LANGUAGE MODELS
- IAG: Induction-Augmented Generation Framework for Answering Reasoning Questions
- T2Ranking: A large-scale Chinese Benchmark for Passage Ranking
- Factuality Enhanced Language Models for Open-Ended Text Generation
- FRESHLLMS: REFRESHING LARGE LANGUAGE MODELS WITH SEARCH ENGINE AUGMENTATION
- KwaiAgents: Generalized Information-seeking Agent System with Large Language Models
- Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
- Complex Claim Verification with Evidence Retrieved in the Wild
- Retrieval-Augmented Generation for Large Language Models: A Survey
- Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy
- ChatQA: Building GPT-4 Level Conversational QA Models
- RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture
- Benchmarking Large Language Models in Retrieval-Augmented Generation
- HyDE：Precise Zero-Shot Dense Retrieval without Relevance Labels
- PROMPTAGATOR : FEW-SHOT DENSE RETRIEVAL FROM 8 EXAMPLES
- SYNERGISTIC INTERPLAY BETWEEN SEARCH AND LARGE LANGUAGE MODELS FOR INFORMATION RETRIEVAL
- T-RAG: Lessons from the LLM Trenches
- RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation
- ASK THE RIGHT QUESTIONS:ACTIVE QUESTION REFORMULATION WITH REINFORCEMENT LEARNING [传统方案参考]
- Query Expansion Techniques for Information Retrieval a Survey [传统方案参考]
- Learning to Rewrite Queries [传统方案参考]
- Managing Diversity in Airbnb Search[传统方案参考]
- 新向量模型用于Recall和Ranking
- BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
- 网易为RAG设计的BCE Embedding技术报告
LLM+KG
- 综述类
- Unifying Large Language Models and Knowledge Graphs: A Roadmap
- Large Language Models and Knowledge Graphs: Opportunities and Challenges
- 知识图谱与大模型融合实践研究报告2023
- KG用于大模型推理
- Using Large Language Models for Zero-Shot Natural Language Generation from Knowledge Graphs
- MindMap: Knowledge Graph Prompting Sparks Graph of Thoughts in Large Language Models
- Knowledge-Augmented Language Model Prompting for Zero-Shot Knowledge Graph Question Answering
- Domain Specific Question Answering Over Knowledge Graphs Using Logical Programming and Large Language Models
- BRING YOUR OWN KG: Self-Supervised Program Synthesis for Zero-Shot KGQA
- StructGPT: A General Framework for Large Language Model to Reason over Structured Data
- 大模型用于KG构建
- Enhancing Knowledge Graph Construction Using Large Language Models
- LLM-assisted Knowledge Graph Engineering: Experiments with ChatGPT
- ITERATIVE ZERO-SHOT LLM PROMPTING FOR KNOWLEDGE GRAPH CONSTRUCTION
- Exploring Large Language Models for Knowledge Graph Completion
Humanoid Agents
- HABITAT 3.0: A CO-HABITAT FOR HUMANS, AVATARS AND ROBOTS
- Humanoid Agents: Platform for Simulating Human-like Generative Agents
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- Shaping the future of advanced robotics
- AUTORT: EMBODIED FOUNDATION MODELS FOR LARGE SCALE ORCHESTRATION OF ROBOTIC AGENTS
- ROBOTIC TASK GENERALIZATION VIA HINDSIGHT TRAJECTORY SKETCHES
预训练数据(pretrain_data)
- DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
- The Pile: An 800GB Dataset of Diverse Text for Language Modeling
- CCNet: Extracting High Quality Monolingual Datasets fromWeb Crawl Data
- WanJuan: A Comprehensive Multimodal Dataset for Advancing English and Chinese Large Models
- CLUECorpus2020: A Large-scale Chinese Corpus for Pre-training Language Model
- In-Context Pretraining: Language Modeling Beyond Document Boundaries
领域模型 (domain_llms)
- MedGPT: Medical Concept Prediction from Clinical Narratives
- BioGPT：Generative Pre-trained Transformer for Biomedical Text Generation and Mining
- Galactia：A Large Language Model for Science
- PubMed GPT: A Domain-specific large language model for biomedical text ⭐️
- BloombergGPT： A Large Language Model for Finance
- ChatDoctor：Medical Chat Model Fine-tuned on LLaMA Model using Medical Domain Knowledge
- Med-PaLM：Large Language Models Encode Clinical Knowledge[V1,V2] ⭐️
- Augmented Large Language Models with Parametric Knowledge Guiding
- XuanYuan 2.0: A Large Chinese Financial Chat Model with Hundreds of Billions Parameters
- ChatLaw Open-Source Legal Large Language Model ⭐️
- MediaGPT : A Large Language Model For Chinese Media
- SMILE: Single-turn to Multi-turn Inclusive Language Expansion via ChatGPT for Mental Health Support
- KITLM: Domain-Specific Knowledge InTegration into Language Models for Question Answering
- FinVis-GPT: A Multimodal Large Language Model for Financial Chart Analysis
- EcomGPT: Instruction-tuning Large Language Models with Chain-of-Task Tasks for E-commerce
- FinGPT: Open-Source Financial Large Language Models
- TableGPT: Towards Unifying Tables, Nature Language and Commands into One GPT
- CFGPT: Chinese Financial Assistant with Large Language Model
- Zhongjing: Enhancing the Chinese Medical Capabilities of Large Language Model through Expert Feedback and Real-world Multi-turn Dialogue
- LLEMMA: AN OPEN LANGUAGE MODEL FOR MATHEMATICS
- CFBenchmark: Chinese Financial Assistant Benchmark for Large Language Model
- InvestLM: A Large Language Model for Investment using Financial Domain Instruction Tuning
- WeaverBird: Empowering Financial Decision-Making with Large Language Model, Knowledge Base, and Search Engine
- FinMem: A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design
- MEDITAB: SCALING MEDICAL TABULAR DATA PREDICTORS VIA DATA CONSOLIDATION, ENRICHMENT, AND REFINEMENT
- PLLaMa: An Open-source Large Language Model for Plant Science
LLM超长文本处理 (long_input)
- 位置编码、注意力机制优化
- Unlimiformer: Long-Range Transformers with Unlimited Length Input
- Parallel Context Windows for Large Language Models
- 苏剑林, NBCE：使用朴素贝叶斯扩展LLM的Context处理长度 ⭐️
- Structured Prompting: Scaling In-Context Learning to 1,000 Examples
- Vcc: Scaling Transformers to 128K Tokens or More by Prioritizing Important Tokens
- Scaling Transformer to 1M tokens and beyond with RMT
- TRAIN SHORT, TEST LONG: ATTENTION WITH LINEAR BIASES ENABLES INPUT LENGTH EXTRAPOLATION ⭐️
- Extending Context Window of Large Language Models via Positional Interpolation
- LongNet: Scaling Transformers to 1,000,000,000 Tokens
- https://kaiokendev.github.io/til#extending-context-to-8k
- 苏剑林,Transformer升级之路：10、RoPE是一种β进制编码 ⭐️
- 苏剑林,Transformer升级之路：11、将β进制位置进行到底
- 苏剑林,Transformer升级之路：12、无限外推的ReRoPE？
- 苏剑林,Transformer升级之路：15、Key归一化助力长度外推
- EFFICIENT STREAMING LANGUAGE MODELS WITH ATTENTION SINKS
- Ring Attention with Blockwise Transformers for Near-Infinite Context
- YaRN: Efficient Context Window Extension of Large Language Models
- LM-INFINITE: SIMPLE ON-THE-FLY LENGTH GENERALIZATION FOR LARGE LANGUAGE MODELS
- EFFICIENT STREAMING LANGUAGE MODELS WITH ATTENTION SINKS
- 上文压缩排序方案
- Lost in the Middle: How Language Models Use Long Contexts ⭐️
- LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models
- LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression ⭐️
- Learning to Compress Prompts with Gist Tokens
- Unlocking Context Constraints of LLMs: Enhancing Context Efficiency of LLMs with Self-Information-Based Content Filtering
- LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
- 训练和模型架构方案
- Never Train from Scratch: FAIR COMPARISON OF LONGSEQUENCE MODELS REQUIRES DATA-DRIVEN PRIORS
- Soaring from 4K to 400K: Extending LLM’s Context with Activation Beacon
- Never Lost in the Middle: Improving Large Language Models via Attention Strengthening Question Answering
- Focused Transformer: Contrastive Training for Context Scaling
- Effective Long-Context Scaling of Foundation Models
- ON THE LONG RANGE ABILITIES OF TRANSFORMERS
- Efficient Long-Range Transformers: You Need to Attend More, but Not Necessarily at Every Layer
- POSE: EFFICIENT CONTEXT WINDOW EXTENSION OF LLMS VIA POSITIONAL SKIP-WISE TRAINING
- LONGLORA: EFFICIENT FINE-TUNING OF LONGCONTEXT LARGE LANGUAGE MODELS
- LongAlign: A Recipe for Long Context Alignment of Large Language Models
- Data Engineering for Scaling Language Models to 128K Context
- 效率优化
- Efficient Attention: Attention with Linear Complexities
- Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
- HyperAttention: Long-context Attention in Near-Linear Time
- FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
- With Greater Text Comes Greater Necessity: Inference-Time Training Helps Long Text Generation
LLM长文本生成（long_output）
- Re3 : Generating Longer Stories With Recursive Reprompting and Revision
- RECURRENTGPT: Interactive Generation of (Arbitrarily) Long Text
- DOC: Improving Long Story Coherence With Detailed Outline Control
- Weaver: Foundation Models for Creative Writing
- Assisting in Writing Wikipedia-like Articles From Scratch with Large Language Models
NL2SQL
- 大模型方案
- DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correction ⭐️
- C3: Zero-shot Text-to-SQL with ChatGPT ⭐️
- SQL-PALM: IMPROVED LARGE LANGUAGE MODEL ADAPTATION FOR TEXT-TO-SQL
- BIRD Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQL ⭐️
- A Case-Based Reasoning Framework for Adaptive Prompting in Cross-Domain Text-to-SQL
- ChatDB: AUGMENTING LLMS WITH DATABASES AS THEIR SYMBOLIC MEMORY
- A comprehensive evaluation of ChatGPT’s zero-shot Text-to-SQL capability
- Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning
- Domain Knowledge Intensive
- Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge
- Bridging the Generalization Gap in Text-to-SQL Parsing with Schema Expansion
- Towards Robustness of Text-to-SQL Models against Synonym Substitution
- FinQA: A Dataset of Numerical Reasoning over Financial Data
- others
- RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL
- MIGA: A Unified Multi-task Generation Framework for Conversational Text-to-SQL
Code Generation
- Code Generation with AlphaCodium: From Prompt Engineering to Flow Engineering
- Codeforces as an Educational Platform for Learning Programming in Digitalization
- Competition-Level Code Generation with AlphaCode
- CODECHAIN: TOWARDS MODULAR CODE GENERATION THROUGH CHAIN OF SELF-REVISIONS WITH REPRESENTATIVE SUB-MODULES
降低模型幻觉 (reliability)
- Survey
- Large language models and the perils of their hallucinations
- Survey of Hallucination in Natural Language Generation
- Siren’s Song in the AI Ocean: A Survey on Hallucination in Large Language Models
- A Survey of Hallucination in Large Foundation Models
- A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
- Calibrated Language Models Must Hallucinate
- Why Does ChatGPT Fall Short in Providing Truthful Answers?
- Prompt or Tunning
- R-Tuning: Teaching Large Language Models to Refuse Unknown Questions
- PROMPTING GPT-3 TO BE RELIABLE
- ASK ME ANYTHING: A SIMPLE STRATEGY FOR PROMPTING LANGUAGE MODELS ⭐️
- On the Advance of Making Language Models Better Reasoners
- RefGPT: Reference → Truthful & Customized Dialogues Generation by GPTs and for GPTs
- Rethinking with Retrieval: Faithful Large Language Model Inference
- GENERATE RATHER THAN RETRIEVE: LARGE LANGUAGE MODELS ARE STRONG CONTEXT GENERATORS
- Large Language Models Struggle to Learn Long-Tail Knowledge
- Decoding Strategy
- Trusting Your Evidence: Hallucinate Less with Context-aware Decoding ⭐️
- SELF-REFINE:ITERATIVE REFINEMENT WITH SELF-FEEDBACK ⭐️
- Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference
- Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
- Enabling Large Language Models to Generate Text with Citations
- Factuality Enhanced Language Models for Open-Ended Text Generation
- KL-Divergence Guided Temperature Sampling
- KCTS: Knowledge-Constrained Tree Search Decoding with Token-Level Hallucination Detection
- CONTRASTIVE DECODING IMPROVES REASONING IN LARGE LANGUAGE MODEL
- Contrastive Decoding: Open-ended Text Generation as Optimization
- Probing and Detection
- Automatic Evaluation of Attribution by Large Language Models
- QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarization
- Zero-Resource Hallucination Prevention for Large Language Models
- LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examples
- Language Models (Mostly) Know What They Know ⭐️
- LM vs LM: Detecting Factual Errors via Cross Examination
- Do Language Models Know When They’re Hallucinating References?
- SELFCHECKGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models
- SELF-CONTRADICTORY HALLUCINATIONS OF LLMS: EVALUATION, DETECTION AND MITIGATION
- Self-consistency for open-ended generations
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- Selective-LAMA: Selective Prediction for Confidence-Aware Evaluation of Language Models
- Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs
- Reviewing and Calibration
- Truth-o-meter: Collaborating with llm in fighting its hallucinations
- RARR: Researching and Revising What Language Models Say, Using Language Models
- CRITIC: LARGE LANGUAGE MODELS CAN SELFCORRECT WITH TOOL-INTERACTIVE CRITIQUING
- VALIDATING LARGE LANGUAGE MODELS WITH RELM
- PURR: Efficiently Editing Language Model Hallucinations by Denoising Language Model Corruptions
- Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback
- Adaptive Chameleon or Stubborn Sloth: Unraveling the Behavior of Large Language Models in Knowledge Clashes
- Woodpecker: Hallucination Correction for Multimodal Large Language Models
- Zero-shot Faithful Factual Error Correction
大模型评估（evaluation）
- 事实性评估
- TRUSTWORTHY LLMS: A SURVEY AND GUIDELINE FOR EVALUATING LARGE LANGUAGE MODELS’ ALIGNMENT
- TrueTeacher: Learning Factual Consistency Evaluation with Large Language Models
- TRUE: Re-evaluating Factual Consistency Evaluation
- FACTSCORE: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation
- KoLA: Carefully Benchmarking World Knowledge of Large Language Models
- When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories
- FACTOOL: Factuality Detection in Generative AI A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
- 检测任务
- Detecting Pretraining Data from Large Language Models
- Scalable Extraction of Training Data from (Production) Language Models
- Rethinking Benchmark and Contamination for Language Models with Rephrased Samples
推理优化(inference)
- Fast Transformer Decoding: One Write-Head is All You Need
- Fast Inference from Transformers via Speculative Decoding
- GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
- Skeleton-of-Thought: Large Language Models Can Do Parallel Decoding
- SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference
- BatchPrompt: Accomplish more with less
模型知识编辑黑科技(model_edit)
- ROME：Locating and Editing Factual Associations in GPT
- Transformer Feed-Forward Layers Are Key-Value Memories
- MEMIT: Mass-Editing Memory in a Transformer
- MEND：Fast Model Editing at Scale
- Editing Large Language Models: Problems, Methods, and Opportunities
- Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
模型合并和剪枝(model_merge)
- Blending Is All You Need: Cheaper, Better Alternative to Trillion-Parameters LLM
- DARE Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
- EDITING MODELS WITH TASK ARITHMETIC
- TIES-Merging: Resolving Interference When Merging Models
- LM-Cocktail: Resilient Tuning of Language Models via Model Merging
- SLICEGPT: COMPRESS LARGE LANGUAGE MODELS BY DELETING ROWS AND COLUMNS
Other Prompt Engineer(prompt_engineer)
- Calibrate Before Use: Improving Few-Shot Performance of Language Models
- In-Context Instruction Learning
- LEARNING PERFORMANCE-IMPROVING CODE EDITS
- Boosting Theory-of-Mind Performance in Large Language Models via Prompting
- Generated Knowledge Prompting for Commonsense Reasoning
- RECITATION-AUGMENTED LANGUAGE MODELS
- kNN PROMPTING: BEYOND-CONTEXT LEARNING WITH CALIBRATION-FREE NEAREST NEIGHBOR INFERENCE
- EmotionPrompt: Leveraging Psychology for Large Language Models Enhancement via Emotional Stimulus
- Causality-aware Concept Extraction based on Knowledge-guided Prompting
- LARGE LANGUAGE MODELS AS OPTIMIZERS
Multimodal
- InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning
- Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models
- PaLM-E: An Embodied Multimodal Language Model
- LLava Visual Instruction Tuning
- MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models
- TabLLM: Few-shot Classification of Tabular Data with Large Language Models
- BLIVA: A Simple Multimodal LLM for Better Handling of Text-Rich Visual Questions
- mPLUG-Owl : Modularization Empowers Large Language Models with Multimodality
- LVLM eHub: A Comprehensive Evaluation Benchmark for Large VisionLanguage Models
- Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities
- Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models
- AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling
- Sora tech report
Timeseries LLM
- TimeGPT-1
- Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook
- TIME-LLM: TIME SERIES FORECASTING BY REPROGRAMMING LARGE LANGUAGE MODELS
- Large Language Models Are Zero-Shot Time Series Forecasters
- TEMPO: PROMPT-BASED GENERATIVE PRE-TRAINED TRANSFORMER FOR TIME SERIES FORECASTING
- Generative Pre-Training of Time-Series Data for Unsupervised Fault Detection in Semiconductor Manufacturing
- Lag-Llama: Towards Foundation Models for Time Series Forecasting
- PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecasting
Quanization
- AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
- LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
- LLM.int8() 8-bit Matrix Multiplication for Transformers at Scale
- SmoothQuant Accurate and Efficient Post-Training Quantization for Large Language Models
Adversarial Attacking
- Curiosity-driven Red-teaming for Large Language Models
- Red Teaming Language Models with Language Models
- EXPLORE, ESTABLISH, EXPLOIT: RED-TEAMING LANGUAGE MODELS FROM SCRATCH
Others
- Pretraining on the Test Set Is All You Need 哈哈作者你是懂讽刺文学的
- Learnware: Small Models Do Big
- The economic potential of generative AI
- A PhD Student’s Perspective on Research in NLP in the Era of Very Large Language Models
近年来，人工智能领域的发展迅猛，特别是在自然语言处理（NLP）领域，大型语言模型（LLM）和提示工程（Prompt Engineering）已经成为研究的热点。同时，开源数据集和模型的共享，以及AI生成内容（AIGC）的应用也在不断扩展。以下是对这些领域的一个简要总结：
Prompt Engineering（提示工程）
Prompt Engineering 是指设计和优化输入提示（prompts）以引导大型语言模型（LLM）生成特定输出的技术。这种方法利用了LLM在预训练阶段学习到的丰富知识，通过精心设计的提示来激发模型的潜能，从而在各种任务上实现零样本（zero-shot）或少样本（few-shot）学习。
关键点：
- 零样本学习：不依赖于特定任务的训练数据，仅通过提示来指导模型完成任务。
- 少样本学习：使用少量标注数据进行快速调整，以适应新任务。
- 提示设计：选择合适的提示对于提高模型性能至关重要，包括手工设计的固定提示和通过算法生成的动态提示。
开源数据&模型
开源数据集和模型的共享对于推动AI研究和应用至关重要。开源数据集提供了丰富的、多样化的数据资源，而开源模型则允许研究者和开发者在已有的基础上进行创新和改进。
关键点：
- 数据集：高质量的开源数据集，如WikiText、SQuAD等，为模型训练提供了基础。
- 预训练模型：大型预训练模型，如BERT、GPT-3等，通过在海量数据上的训练，学习到了丰富的语言表示。
- 社区贡献：研究者和开发者通过贡献数据集、模型和工具，共同推动了AI技术的进步。
AIGC（AI Generated Content）
AIGC 指的是使用人工智能技术自动生成内容，这包括文本、图像、音频和视频等。AIGC 应用的兴起，使得个性化内容的创作变得更加高效和便捷。
关键点：
- 内容创作：AIGC可以用于新闻撰写、小说创作、艺术作品生成等。
- 个性化定制：根据用户的偏好和需求，生成定制化的内容。
- 版权和伦理问题：AIGC 带来了版权、原创性和伦理等方面的挑战。
总结
Prompt Engineering、开源数据&模型以及AIGC 应用是人工智能领域相互关联的三个重要方面。Prompt Engineering 通过优化输入提示来提升LLM的性能，开源数据和模型的共享为AI研究提供了基础，而AIGC 应用则展示了AI技术在内容创作领域的潜力。随着技术的进步，这些领域将继续发展，为人类社会带来更多的创新和变革。