demilich

7 ms·

1.
by demilich 2y ago
https://github.com/infiniflow/infinity , dense vector + sparse vector + fulltext search(BM25) + late interact reranker(Colbert)
2.
by demilich 2y ago
GraphRAG is to parse data to create a KG and retrieval the information from KB. RAGFlow is to create a Graph workflow to solve multi-hop question-answering issue.
3.
by demilich 2y ago
Current RAG is more and more complex, will future LLM or New AI model overthrow the RAG?
4.
Nemotron-4 340B(research.nvidia.com)
3 points by demilich 2y ago|0 comments
5.
by demilich 2y ago
Infinity supports HNSW vector index.
6.
by demilich 2y ago
Check https://github.com/infiniflow/infinity which combines vector search and full-text search providing extremely fast search performance.
7.
by demilich 2y ago
Try RAPTOR: https://arxiv.org/html/2401.18059v1 An implementation: github.com/infiniflow/ragflow
8.
by demilich 2y ago
Agreed
9.
by demilich 2y ago
RAGFlow (github.com/infiniflow/ragflow) use OCR/layout recognition/TSR(table structure recognition) to understand the document structure and context. Is there any difference between RAGFlow and ZenDB?
10.
by demilich 2y ago
Use C++20 modules, take a look at this project: https://github.com/infiniflow/infinity
11.
by demilich 2y ago
Cool! This is really helpful.
12.
by demilich 2y ago
Each project has its own detailed requirements and scenarios, and we cannot demand that each project use same library to implement similar functions
13.
by demilich 2y ago
The recognition of file layout to parse file content is indeed a very innovative idea. Compared to many open-source projects we have seen before, this provides us with new ideas for using RAG to solve problems in the future. I hope the auth
14.
by demilich 6y ago
Well, I suppose its website have answered the similar question. From my point of view, FAISS and ANNOY are algorithm libraries which can be integrated into any application, but can't be used directly. Meanwhile, Milvus is a server can
15.
by demilich 6y ago
Good job, excited!