LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget
Paper • 2607.14952 • Published • 210
OCR
State-of-the-art OCR models, datasets, and demos for document parsing.
-
-
VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding
Paper • 2607.14935 • Published • 172 -
From Pixels to States: Rethinking Interactive World Models as Game Engines
Paper • 2607.14076 • Published • 36 -
baidu/Unlimited-OCR
Image-Text-to-Text • 3B • Updated • 2.84M • 3.95k -
zai-org/GLM-OCR
Image-Text-to-Text • 1B • Updated • 3.59M • • 1.97k -
nvidia/nemotron-ocr-v2
Image-to-Text • Updated • 4k • 240 -
uv-scripts/ocr
Updated • 3.18k • 155 -
bevaya/pubmed-ocr
Viewer • Updated • 1.55M • 3.08k • 72 -
unsloth/LaTeX_OCR
Viewer • Updated • 76.3k • 3.48k • 90