[RFC] 061 - Multiple RAG Strategies Support #4204
cy948
started this conversation in
RFC | 特性开发
Replies: 1 comment
-
希望可以支持text-embeddings-inference 的 embedding和rerank |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
背景
RAG被广泛应用于增强LLM在没有内部知识支撑的场景下的性能表现。如领域知识敏感的任务1、幻觉问题的解决2 。近年来对RAG的Pipeline研究主要聚焦于对三个核心模块:Retrieval, Augment, Generation3的性能提升。因此,本RFC目标有:
思路
2024/10/6
进展
2024/9/29
: RFC初步提出、方案实验;2024/10/13
: 找到后端rag策略定义,准备开始改造lobe-chat/src/server/routers/lambda/chunk.ts
Line 129 in 9a369ac
2024/10/16
: 在maintainer的建议下,先然所接入类似功能的 Dify以实现对 Agent 架构的支持。 [RFC] 064 - Dify Integration | Dify 接入 #4412Reference
1 N. Kandpal, H. Deng, A. Roberts, E. Wallace, and C. Raffel, “Large language models struggle to learn long-tail knowledge,” in International Conference on Machine Learning. PMLR, 2023, pp. 15 696–15 707.
2 Y. Zhang, Y. Li, L. Cui, D. Cai, L. Liu, T. Fu, X. Huang, E. Zhao, Y. Zhang, Y. Chen et al., “Siren’s song in the ai ocean: A survey on hallucination in large language models,” arXiv preprint arXiv:2309.01219, 2023.
3 Gao Y, Xiong Y, Gao X, et al. Retrieval-augmented generation for large language models: A survey[J]. arXiv preprint arXiv:2312.10997, 2023.
Issues
Beta Was this translation helpful? Give feedback.
All reactions