Sasha Rush

@srush_nlp

Lean Verified Tokenizers (https://srush.github.io/verified-tokenizer/) Training-inference mismatch kills RL runs, but yet we build messages in Jinja? Prototypes a Lean-verified Rust agent pipeline.
打开原帖#511482
  1. Industry

    Mistral AI: Served in the EU and US on Mistral AI Cloud Inference
  2. Industry

    Mistral AI: We’re also now serving two fast, low-cost models: - DeepSeek V4.1 Fla…
  3. Industry

    Mistral AI: Today, we’re introducing the Responses API in public preview, made fo…