AI & ML interests
None defined yet.
Recent Activity
Papers
ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models
LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR
Articles
mDenseOn & mLateOn pareto frontiers
Explore model size vs. retrieval quality with interactive charts
LightOn-rerank-LW
Compare and rerank four text passages or document pages
LightOn-rerank-PW
LightOn multimodal document reranker with pointwise scoring
HPool Results Dashboard
Explore BEIR evaluation results with interactive tables
LightOnOCR 2 1B Demo
Extract text and layout from images or PDF documents
LightOnOCR 1B Demo
Extract text from images or PDFs with OCR
OriOn Leaderboard
Leaderboard for all OriOn training runs for reproducibility!
LightOnOCR 1B Demo Zero
Extract text from images and PDFs
LightOnOCR 2 1B Examples
Examples of OCR performance for LigthOnOCR-2-1B models
LightOnOCR 2 1B Examples
Examples of OCR performance for LigthOnOCR-2-1B models
MCPyLate
MCP server to perform search using PyLate
Vision Pipeline
Answer questions about uploaded images and documents