Having worked on unlearning for multiple years, it was clear that post-training "fixes" alone were a dead-end. Moโ€ฆ

Having worked on unlearning for multiple years, it was clear that post-training "fixes" alone were a dead-end. Model learning is way too entangled. With ๐—ก๐—จ๐—Ÿ๐—Ÿ๐˜€ we decided to architect unlearnability into the model, and scaled it to 1B+
Ranked #79 on backlist 2026-06-17 (17 Jun 2026 UTC) ยท by (Pratyush Maini) ยท

How it ranks: Backlist reads my Twitter/X timeline, scores every tweet for substance with an LLM rubric (not engagement), and publishes the daily top picks with a one-line takeaway. Curated by Surya Dantuluri.