---
title: MTLA
emoji: 🧠
colorFrom: yellow
colorTo: indigo
sdk: static
pinned: false
license: apache-2.0
short_description: 'Propose and Attend: Training-free MLLM Grounding Confidence'
arxiv: 2607.05978
---

# Propose and Attend: Training-free MLLM Grounding Confidence via Multi-Token Localized Attention

This is the Hugging Face Space project page for **Propose and Attend (MTLA)**.

*   **Paper (arXiv):** [https://arxiv.org/abs/2607.05978](https://arxiv.org/abs/2607.05978)
*   **GitHub Repository:** [https://github.com/TalRemez/MTLA](https://github.com/TalRemez/MTLA)

## Citation
```bibtex
@article{shalam2026propose,
  title={Propose and Attend: Training-free MLLM Grounding Confidence via Multi-Token Localized Attention},
  author={Shalam, Daniel and Baruch, Emanuel Ben and Cohen, Avi Ben and Remez, Tal},
  journal={arXiv preprint arXiv:2607.05978},
  year={2026}
}
```