The results are simply impressive.
LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR by Said Taghadouini, Adrien Cavaillès, Baptiste Aubertin
We present \textbf{LightOnOCR-2-1B}, a 1B-parameter end-to-end multilingual vision--language model that converts document images (e.g., PDFs) into clean, naturally ordered text without brittle OCR pipelines. Trained on a large-scale, high-quality distillation mix with strong coverage of scans, French documents, and scientific PDFs, LightOnOCR-2 achieves state-of-the-art results on OlmOCR-Bench while being 9\times smaller and substantially faster than prior best-performing models. We further extend the output format to predict normalized bounding boxes for embedded images, introducing localization during pretraining via a resume strategy and refining it with RLVR using IoU-based rewards. Finally, we improve robustness with checkpoint averaging and task-arithmetic merging. We release model checkpoints under Apache 2.0, and publicly release the dataset and \textbf{LightOnOCR-bbox-bench} evaluation under their respective licenses.
More information on LightOn's blog.
Paris Machine Learning: Meetup.com||@Archives||LinkedIn||Facebook|| @ParisMLGroup About LightOn: Newsletter ||@LightOnIO|| on LinkedIn || on CrunchBase || our Blog
About myself: LightOn || Google Scholar || LinkedIn ||@IgorCarron ||Homepage||ArXiv
No comments:
Post a Comment