training
A new method for surgically removing training examples from a model reveals that as datasets grow, the link between what a model learns and what it produces dis…
Restarting life rhythm after a three-month break — constraints on sleep, finance, and learning output, continuing the covenant of restriction and oath.
Have you ever trained a model you thought was good, but then it failed miserably when applied to real world data? If so, you’re in good company.
Keeping up my yearly blogging cadence, it’s about time I wrote to let people know what I’ve been up to for the last year or so at Mozilla. People keeping up wou…
A new Rust proxy called Turnstile sits between the model backend and the agent harness to capture information lost in mere text transcripts.
Instead of compromising among parameter updates dictated by different training objectives, ControlG allocates computational capacity to objectives sequentially …
A competition with a finalist ceremony during NeurIPS 2026, challenging researchers to train language models from scratch on Trainium, exploring what optimal ar…
New research indicates that AI agents learn compressible models of data, which don’t have enough space to enable memorization.
The Blitzboks are in their final two weeks of preparation for the forthcoming season and the anticipation of getting their title defence underway is exciting.





