Exploratory Vision-RAG lab for page-level document retrieval and multimodal PDF understanding.
-
Updated
May 21, 2026 - Jupyter Notebook
#
colqwen
Here are 3 public repositories matching this topic...
Interactive playground for configuring, benchmarking, and debugging Docling document parsing pipelines and test the quality of late retrieval.
-
Updated
Oct 16, 2026 - TypeScript
EPFL LiGHT lab × Télécom SudParis (Cassiopée project) — benchmark of six ColVision visual retrievers (ColPali, ColQwen2/2.5, ColGemma3, ColSmol) against text retrieval inside the mmore RAG pipeline, on ViDoRe v2 and native-language biomedical literature in five languages.
multilingual benchmark information-retrieval epfl document-retrieval rag telecom-sudparis vision-language-models multimodal-rag colpali colqwen mmore
-
Updated
Oct 22, 2026 - Python
Add this topic to your repo
To associate your repository with the colqwen topic, visit your repo's landing page and select "manage topics."