lfoppiano commited on
Commit
13f46f5
1 Parent(s): 99f35d8

documentation

Browse files
Files changed (1) hide show
  1. README.md +7 -4
README.md CHANGED
@@ -16,11 +16,14 @@ license: apache-2.0
16
 
17
  ## Introduction
18
 
19
- Question/Answering on scientific documents using LLMs (OpenAI, Mistral, ~~LLama2,~~ etc..).
20
- This application is the frontend for testing the RAG (Retrieval Augmented Generation) on scientific documents, that we are developing at NIMS.
21
- Differently to most of the project, we focus on scientific articles. We target only the full-text using [Grobid](https://github.com/kermitt2/grobid) that provide and cleaner results than the raw PDF2Text converter (which is comparable with most of other solutions).
 
22
 
23
- **NER in LLM response**: The responses from the LLMs are post-processed to extract <span stype="color:yellow">physical quantities, measurements</span> (with [grobid-quantities](https://github.com/kermitt2/grobid-quantities)) and <span stype="color:blue">materials</span> mentions (with [grobid-superconductors](https://github.com/lfoppiano/grobid-superconductors)).
 
 
24
 
25
  **Demos**:
26
  - (on HuggingFace spaces): https://lfoppiano-document-qa.hf.space/
 
16
 
17
  ## Introduction
18
 
19
+ Question/Answering on scientific documents using LLMs: ChatGPT-3.5-turbo, Mistral-7b-instruct and Zephyr-7b-beta.
20
+ The streamlit application demonstrate the implementaiton of a RAG (Retrieval Augmented Generation) on scientific documents, that we are developing at NIMS (National Institute for Materials Science), in Tsukuba, Japan.
21
+ Differently to most of the projects, we focus on scientific articles.
22
+ We target only the full-text using [Grobid](https://github.com/kermitt2/grobid) that provide and cleaner results than the raw PDF2Text converter (which is comparable with most of other solutions).
23
 
24
+ Additionally, this frontend provides the visualisation of named entities on LLM responses to extract <span stype="color:yellow">physical quantities, measurements</span> (with [grobid-quantities](https://github.com/kermitt2/grobid-quantities)) and <span stype="color:blue">materials</span> mentions (with [grobid-superconductors](https://github.com/lfoppiano/grobid-superconductors)).
25
+
26
+ The conversation is backed up by a sliding window memory (top 4 more recent messages) that help refers to information previously discussed in the chat.
27
 
28
  **Demos**:
29
  - (on HuggingFace spaces): https://lfoppiano-document-qa.hf.space/