〔探索AI的无限可能,微信关注“AIGCmagic”公众号,让AIGC科技点亮生活〕
一. 大模型自我纠错相关资源
https://github.com/ryokamoi/llm-self-correction-papers
自我纠错综述:《When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs》
论文链接:https://arxiv.org/abs/2406.01297
涵盖内在自我纠错、外部工具辅助纠错、信息检索辅助纠错等。
二. OpenAI O1相关工作
https://platform.openai.com/docs/guides/reasoning
1. OpenAI o1
Learning to Reason with LLMs
https://arxiv.org/abs/2409.18486
https://openai.com/index/learning-to-reason-with-llms/
2. Skywork-o1
https://huggingface.co/Skywork
https://huggingface.co/Skywork/Skywork-o1-Open-Llama-3.1-8B
3. LLaVA-CoT
LLaVA-CoT: Let Vision Language Models Reason Step-by-Step
https://arxiv.org/abs/2411.10440
4. Marco-o1
Towards Open Reasoning Models for Open-Ended Solutions
https://arxiv.org/abs/2411.14405
5、QwQ
Reflect Deeply on the Boundaries of the Unknown
https://qwenlm.github.io/blog/qwq-32b-preview/
https://huggingface.co/Qwen/QwQ-32B-Preview
6. Steiner
https://medium.com/@peakji/a-small-step-towards-reproducing-openai-o1-b9a756a00855
https://huggingface.co/collections/peakji/steiner-preview-6712c6987110ce932a44e9a6
7. DeepSeek-R1-Lite-Preview
https://api-docs.deepseek.com/news/news1120
8. O1 Replication Journey
https://github.com/GAIR-NLP/O1-Journey
9. openai-o1 推理prompt
https://openai.com/index/learning-to-reason-with-llms/
https://github.com/bradhilton/o1-chain-of-thought/tree/main/examples
10. COT数据合成方案
g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains
https://github.com/bklieger-groq/g1
multi1: Using multiple AI providers to create o1-like reasoning chains
https://github.com/tcsenpai/multi1
推荐阅读
社区简介:
《AIGCmagic星球》,五大AIGC方向正式上线!让我们在AIGC时代携手同行!限量活动中!
AI多模态核心架构五部曲:
AI多模态模型架构之模态编码器:图像编码、音频编码、视频编码
AI多模态模型架构之输入投影器:LP、MLP和Cross-Attention
AI多模态模型架构之输出映射器:Output Projector
AI多模态模型架构之模态生成器:Modality Generator
AI多模态实战教程:
AI多模态教程:从0到1搭建VisualGLM图文大模型案例