mkd-hika/keural-vision-encoder-mid
Upload README.md with huggingface_hub
Upload README.md with huggingface_hub
Upload README.md with huggingface_hub
Remove redundant Mermaid diagram (superseded by the architecture figure)
Restructure model card: at-a-glance results, working quickstart, GitHub link, reordered sections
Publish final checkpoint at repo root so from_pretrained() works
Publish final checkpoint at repo root so from_pretrained() works
Publish final checkpoint at repo root so from_pretrained() works
Remove pipeline GIF (superseded by the architecture figure)
Replace animated pipeline GIF with the architecture specification figure
Add full architecture specification figure
Rebuild pipeline GIF: fix text overflow, trim dead canvas space, clearer stage-by-stage animation
Remove block diagram reference from model card
Remove architecture block diagram (superseded by the Mermaid diagram)
Add radar + R@5/R@10 charts and architecture block diagram; document that R@1 gap narrows at higher k
Add checkpoint-15000 (step 15000) with alignment sidecar and optimizer state
Add final checkpoint (step 18000, training complete) with alignment sidecar and eval results
Update model card: training complete, full eval results, benchmark comparisons, saliency evolution, loss curves
Replace overflowing ASCII architecture with a native Mermaid block diagram
Professional model card: standard sections, corrected training/architecture facts, intended-use & limitations
Embed animated pipeline GIF in model card
Add animated pipeline preview GIF
Add link to live animated pipeline diagram Space
Upload folder using huggingface_hub
Upload folder using huggingface_hub
Upload README.md with huggingface_hub
initial commit