booydar/t5-experiments

[NeurIPS 22] [AAAI 24] Recurrent Transformer-based long-context architecture.

1 best points · 2 comments · 1 Hacker News post · 777 GitHub stars · View on GitHub

Hacker News discussions

• Code for Scaling Transformer to 1M tokens and beyond with RMT (arxiv.org) (1 points, 2 comments, by mysterybox, 2023-04-24)