blank

Alex Zhang, Sep 26, 2026.

Fig 1.

Modern post-training has become a brute-force paradigm of curating ever more environments and ever longer training horizons.

Alex Zhang, Omar Khattab I believe it’s worth discussing an example of the Mismanaged Geniuses Hypothesis at play: we underestimate how good language models actually are, and they are inhibited by how we use them.

Alex Zhang, Zhening (Zed) Li, and Omar Khattab. tldr; AI models are already good enough for the next leap in capabilities.

Alex Zhang, Feb 25, 2026.

The full paper is now available here: https://www.alphaxiv.org/abs/2512.24601.

This post offers a comprehensive and chronological guide to advances in deep learning from the perspective of efficiency: things like clusters, individual hardware, deep learning libraries, compilers — even architectural changes.

if the LaTeX is not loading, refresh the page.

Introduction To celebrate the end of my graduate school application deadlines and finals week disaster, I decided to spend my winter break going through and reading every single abstract of the accepted papers in the NeurIPS 2023 (which I unfortunately co…