Posts

• Jan 3, 2026

  Label-Supervised Routing in Upcycled Mixture of Experts

• Dec 10, 2025

  A (Somewhat Failed) Experiment in Latent Reasoning with LLMs

• Apr 13, 2025

  translit-former: A Transformer That Speaks Bengali (Kinda)

• Dec 19, 2024

  Watermarking LLM Output

• Jun 30, 2024

  Mixture Density Networks

• Apr 2, 2024

  Mastering Hexapawn AlphaZero-style

subscribe via RSS