A PyTorch Implementation of "Recurrent Models of Visual Attention"
-
Updated
Feb 24, 2023 - Python
A PyTorch Implementation of "Recurrent Models of Visual Attention"
Research on models and agents that improve reasoning and planning through iterative latent computation.
A recurrent Transformer model that reduces transformer parameters by using a single layer in a circular manner, enhanced by adaptive level signals from low-rank matrices. (Findings of the Association for Computational Linguistics: EMNLP 2025)
Weight-tied recurrent message passing for the Graph Tsetlin Machine: one shared clause bank across all rounds, with an honest characterization of when tying helps and when it costs capacity
Repository for COVID-19 screening project. Involves audio processing and some CV.
Studying recurrent transformer blocks for iterative reasoning, error correction, and depth generalization.
To associate your repository with the recurrent-models topic, visit your repo's landing page and select "manage topics."