M3Grounder — Project Page

[CVPR 2026] Official project page for
M3Grounder: Mask-Based Multi-Span and Multi-Granular Grounding for Document QA
BharatGen & IIIT Hyderabad

🔗 Paper · HF Space · Code · Poster

---

What is M3Grounder?

M3Grounder grounds document QA answers at pixel level — not just bounding boxes. It autoregressively generates answer text interleaved with [GROUND] tokens that link each answer span to its evidence region on the page, at phrase, line, and block granularity.

Key ideas:

• Mask-based grounding — fine-grained segmentation instead of coarse boxes
• Multi-span — one answer can cite multiple evidence regions
• Multi-granular — phrase ⊂ line ⊂ block, enforced by an enclosure loss
• Bleed-suppression loss — prevents masks from spilling into irrelevant regions

---

This Repo

Static project page built with plain HTML/CSS/JS. Includes an interactive demo showcasing multi-span, multi-granular grounding across Finance, Legal, and Transcript documents.

index.html          # entire site (single file)
static/
  images/           # logos, teaser, document images

To run locally:

python -m http.server 8000
# open http://localhost:8000

---

Citation

@inproceedings{venna2026m3grounder,
  title     = {M3Grounder: Mask-Based Multi-Span and Multi-Granular Grounding for Document QA},
  author    = {Venna, Venkata Kesav and Gunda, Sai Madhusudan and Jinka, Jyothi Swaroopa
               and Rachakonda, Hrithik Sagar and Srinivasan, Anirudh and Sarvadevabhatla, Ravi Kiran},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages     = {23685--23695},
  year      = {2026}
}