Blog

2026

June 16, 2026 · 7 min read

TokMem: packing a whole procedure into a single token

Compiling each reusable task procedure into a single trainable token, with the language model kept frozen.

June 15, 2026 · 9 min read

ULPT: tuning a soft prompt in two dimensions

Tuning prompts in an ultra-low-dimensional space and letting a frozen random matrix do the rest, for ~98% fewer trainable parameters.

June 14, 2026 · 7 min read

Carrying a soft prompt from one model to another

Describing a soft prompt by what it's near, so its task semantics carry across different language models.