Astra can read and write weights to improve Qwen

Re:Astra can read and write weights to improve Qwen by WarmWash on Monday November 02, 2026 @02:22PM (#50342197)

They are training LLMs to supervise (and improve!) LLMs. This is kind of right out of the playbook.

It's very impressive, but also very expected.

[ Parent | on HN ]

Re:Astra can read and write weights to improve Qwen by hard_times on Monday November 02, 2026 @02:39PM (#50342480)

This I'm afraid is different than the traditional approaches of recursive self-training and self-distillation. Much scarier.

[ Parent | on HN ]