srush pushes Lean-verified tokenizers and formal methods for LLMs

Princeton's Sasha Rush releases Lean-verified Rust tokenizers to fix RL train-inference mismatches, and argues formal verification is the only path to serious LLM-generated code.

2026-10-28 ~ 2026-10-28 · 2 related posts

• srush builds Lean-verified Rust tokenizer pipeline to fix train-inference mismatch that kills RL runs — srush_nlp · 2026-10-28
• srush: formal verification is the only plausible path to serious LLM-written code — srush_nlp · 2026-10-28