{"data":{"slug":"dsxiangli-decryptprompt","name":"DecryptPrompt","tagline":"Summarizes Prompt&LLM Papers, Open-source Data&Models, AIGC Applications","github_url":"https://github.com/DSXiangLi/DecryptPrompt","owner":"DSXiangLi","repo":"DecryptPrompt","owner_avatar_url":"https://avatars.githubusercontent.com/u/37739462?v=4","primary_language":null,"stars":3427,"forks":320,"topics":["aigc","chain-of-thought","chatgpt","demonstration","few-shot-learning","in-context-learning","instruction-tuning","llm","llm-agent","papers","prompt","prompt-engineering","prompt-tuning","zero-shot-learning"],"archived":false,"github_pushed_at":"2026-05-06T00:21:19+00:00","maintenance_label":"Steady","url":"https://www.graphcanon.com/tools/dsxiangli-decryptprompt","markdown_url":"https://www.graphcanon.com/tools/dsxiangli-decryptprompt.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/dsxiangli-decryptprompt","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=dsxiangli-decryptprompt","description":"总结Prompt&LLM论文，开源数据&模型，AIGC应用","homepage_url":null,"license":null,"open_issues":1,"watchers":62,"ai_summary":"DecryptPrompt is an open-source repository summarizing prompt and LLM research papers along with providing associated data sets and models to facilitate understanding and development in AI-driven content generation.","readme_excerpt":"# DecryptPrompt\n> 如果LLM的突然到来让你感到沮丧，不妨读下主目录的Choose Your Weapon Survival Strategies for Depressed AI Academics\n持续更新以下内容，Star to keep updated~\n\n## LLM资源汇总\n- [开源模型和评测榜单](开源模型.MD)\n- [开源推理，微调，Agent，RAG，propmt 框架](开源框架.MD)\n- [开源SFT，RLHF，Pretrain 数据集](开源数据.MD)\n- [AIGC各领域应用汇总](AIGC各领域应用.MD)\n- [Prompt教程，经典博客和AI会议访谈](教程博客会议.MD)\n\n## 跟着博客读论文\n- [解密Prompt系列1. Tunning-Free Prompt：GPT2 & GPT3 & LAMA & AutoPrompt](https://cloud.tencent.com/developer/article/2215545?areaSource=&traceId=)\n- [解密Prompt系列2. 冻结Prompt微调LM： T5 & PET & LM-BFF](https://cloud.tencent.com/developer/article/2223355?areaSource=&traceId=)\n- [解密Prompt系列3. 冻结LM微调Prompt: Prefix-tuning & Prompt-tuning & P-tuning](https://cloud.tencent.com/developer/article/2237259?areaSource=&traceId=)\n- [解密Prompt系列4. 升级Instruction Tuning：Flan/T0/InstructGPT/TKInstruct](https://cloud.tencent.com/developer/article/2245094?areaSource=&traceId=)\n- [解密prompt系列5. APE+SELF=自动化指令集构建代码实现](https://cloud.tencent.com/developer/article/2260697?areaSource=&traceId=)\n- [解密Prompt系列6. lora指令微调扣细节-请冷静,1个小时真不够~](https://cloud.tencent.com/developer/article/2276508)\n- [解密Prompt系列7. 偏好对齐RLHF-OpenAI·DeepMind·Anthropic对比分析](https://cloud.tencent.com/developer/article/old/2289566?areaSource=&traceId=)\n- [解密Prompt系列8. 无需训练让LLM支持超长输入:知识库 & Unlimiformer & PCW & NBCE ](https://cloud.tencent.com/developer/article/old/2295783?areaSource=&traceId=)\n- [解密Prompt系列9. COT：模型复杂推理-思维链基础和进阶玩法](https://cloud.tencent.com/developer/article/old/2296079?areaSource=&traceId=)\n- [解密Prompt系列10. COT：思维链COT原理探究](https://cloud.tencent.com/developer/article/old/2298660)\n- [解密Prompt系列11. COT：小模型也能COT，先天不足后天补](https://cloud.tencent.com/developer/article/old/2301999)\n- [解密Prompt系列12. LLM Agent零微调范式 ReAct & Self Ask](https://cloud.tencent.com/developer/article/2305421)\n- [解密Prompt系列13. LLM Agent指令微调方案: Toolformer & Gorilla](https://cloud.tencent.com/developer/article/2312674)\n- [解密Prompt系列14. LLM Agent之搜索应用设计：WebGPT & WebGLM & WebCPM](https://cloud.tencent.com/developer/article/2319879)\n- [解密Prompt系列15. LLM Agent之数据库应用设计：DIN & C3 & SQL-Palm & BIRD](https://cloud.tencent.com/developer/article/2328749)\n- [解密Prompt系列16. LLM对齐经验之数据越少越好？LTD & LIMA & AlpaGasus](https://cloud.tencent.com/developer/article/2333495)\n- [解密Prompt系列17. LLM对齐方案再升级 WizardLM & BackTranslation & SELF-ALIGN](https://cloud.tencent.com/developer/article/2338592)\n- [解密Prompt系列18. LLM Agent之只有智能体的世界](https://cloud.tencent.com/developer/article/2351540)\n- [解密Prompt系列19. LLM Agent之数据分析领域的应用：Data-Copilot & InsightPilot](https://cloud.tencent.com/developer/article/2358413)\n- [解密Prompt系列20. RAG之再谈召回多样性优化](https://cloud.tencent.com/developer/article/2365050)\n- [解密Prompt系列21. RAG之再谈召回信息密度和质量](https://cloud.tencent.com/developer/article/2369977)\n- [​解密Prompt系列22. RAG的反思：放弃了压缩还是智能么？](https://cloud.tencent.com/developer/article/2375066)\n- [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](https://cloud.tencent.com/developer/article/2378383)\n- [解密prompt系列24. RLHF新方案之训练策略：SLiC-HF & DPO & RRHF & RSO](https://cloud.tencent.com/developer/article/2389619)\n- [解密prompt系列25. RLHF改良方案之样本标注：RLAIF & SALMON](https://cloud.tencent.com/developer/article/2398654)\n- [解密prompt系列26. 人类思考vs模型思考：抽象和发散思维](https://cloud.tencent.com/developer/article/2394120)\n- [解密prompt系列27. LLM对齐经验之如何降低通用能力损失](https://cloud.tencent.com/developer/article/2406888)\n- [解密Prompt系列28. LLM Agent之金融领域智能体：FinMem & FinAgent](https://cloud.tencent.com/developer/article/2411792)\n- [解密Prompt系列29. LLM Agent之真实世界海量API解决方案：ToolLLM & AnyTool](https://cloud.tencent.com/developer/article/2415908)\n- [解密Prompt系列30. LLM Agent之互联网冲浪智能体们](https://cloud.tencent.com/developer/article/2419768)\n- [​解密Prompt系列31. LLM Agent之从经验中不断学习的智能体](https://cloud.tencent.com/developer/article/2425139)\n- [解密Prompt系列32. LLM之表格理解任务-文本模态](https://cloud.tencent.com/developer/article/2429900)\n- [解密Prompt系列33. LLM之图表理解任务-多模态篇](https://cloud.tencent.com/developer/article/2433883)\n- [​解密prompt系列34. RLHF之训练另辟蹊径：循序渐进 & 青出于蓝](https://cloud.tencent.com/developer/article/2437031)\n- [解密prompt系列35. Prom","github_created_at":"2023-02-10T14:10:38+00:00","created_at":"2026-07-11T11:57:57.528963+00:00","updated_at":"2026-07-28T06:00:33.767635+00:00","categories":[{"slug":"developer-tools","name":"Developer Tools","url":"https://www.graphcanon.com/categories/developer-tools","markdown_url":"https://www.graphcanon.com/categories/developer-tools.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/developer-tools"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"aigc","name":"aigc"},{"slug":"chain-of-thought","name":"chain-of-thought"},{"slug":"chatgpt","name":"chatgpt"},{"slug":"demonstration","name":"demonstration"},{"slug":"few-shot-learning","name":"few-shot-learning"},{"slug":"in-context-learning","name":"in-context-learning"},{"slug":"instruction-tuning","name":"instruction-tuning"},{"slug":"llm","name":"llm"}],"trust":{"provenance":{"is_fork":false,"github_id":600068422,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-28T06:00:33.047Z","maintenance":{"label":"Steady","score":60,"methodology":"github_public_v1","releases_90d":0,"days_since_push":83,"last_release_at":null},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:57:58.818Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-07-28T06:00:33.514Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need detailed summaries of prompt-engineering and LLM-related research, as DecryptPrompt is dedicated to this area.","If your project requires access to open-source data sets and models specifically connected with prompt research, given that DecryptPrompt focuses on providing these resources."],"when_not_to_use":["Avoid using DecryptPrompt if you require a solution that supports languages other than English effectively, as the repository's descriptions are in Chinese.","If your development needs go beyond summarization and data/model provision into complex coding examples or comprehensive API documentation, DecryptPrompt might not satisfy these requirements."],"source":"enrich:decision_facts","observed_at":"2026-07-17T12:02:34.667Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"DecryptPrompt is an open-source repository that summarizes prompt and large language model research papers while offering related datasets and models for AI content generation facilitation."}]}}