Anthony Fuller

Thanks for comparing with LookHere. Looking forward to reading this 🙂

Galileo, created by Mila researchers Gabriel Tseng and David Rolnick, uses AI to uncover trends across decades of satellite and human activity data—revealing early signals about our planet’s health and helping us act in time.
mila.quebec/en/article/d...

LookWhere has been accepted to NeurIPS 2025!

LookWhere accelerates inference and fine-tuning by approximating full, deep representations with adaptive computation of predictions learned from distillation.

Paper: arxiv.org/abs/2505.18051

Code and weights: github.com/antofuller/l...

Our remote sensing foundation model Galileo has been accepted to ICML 2025!

Galileo outperforms state-of-the-art across different input data modalities and shapes, and using it requires only minimal data and compute.

More at:
arxiv.org/abs/2502.09356

Elbows up, Canada.

LookHere: Vision Transformers with Directed Attention Generalize and Extrapolate High-resolution images offer more information about scenes that can improve model accuracy. However, the dominant model architecture in computer vision, the vision transformer (ViT), cannot effectivel...

Lookhere position encoding could help with this: arxiv.org/abs/2405.13985

Awesome, thanks for the explanation!

DeBERTa: Decoding-enhanced BERT with Disentangled Attention Recent progress in pre-trained neural language models has significantly improved the performance of many natural language processing (NLP) tasks. In this paper we propose a new model architecture DeBE...

Cool work? I think the position encoding method looks similar to DeBERTa’s disentangled attention: arxiv.org/abs/2006.03654

Writing a good scientific paper

For those who missed this post on the-network-that-is-not-to-be-named, I made public my "secrets" for writing a good CVPR paper (or any scientific paper). I've compiled these tips of many years. It's long but hopefully it helps people write better papers. perceiving-systems.blog/en/post/writ...