From IcePop to KPop — our team keeps pushing on RL training stability for large MoE models. 👇
KPop replaces the fixed-ratio mask with an adaptive binary-KL region that matches each token's inherent noise. More robust updates, stable long-horizon agentic RL.
Ring-2.6-1T → 76+ on SWE-bench Verified, pure RL.
Congrats to @Jia__Guo & team!
Blog:
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
SWE-bench — The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.
mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
Why it matters
tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition →-level mismatch in mixture-of-expertsA model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.Full definition → reinforcement learning is not uniform, and sizing the trust region per token instead of by a fixed ratio is what let a 1T model train stably to 76+ on SWE-benchThe standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.Full definition → with pure RL.
