Qiuyang Mang

@mangqiuyang

PhD student at UC Berkeley @BerkeleySky. Former ICPC World Finalists; Opinions are my own

following
followers

EasyPPO is pitched as more stable PPO for LLM post-training

The EasyPPO announcement says it fixes the critic while leaving the actor update unchanged, with no new actor loss or policy algorithm.

6 Sources, 20d ago, updated 20d ago

Human coders reportedly show super-linear scaling on 14-day tasks

A researcher comparing expert coders with coding agents on AtCoder Heuristic Contest tasks interprets the human results as evidence that people learn about a problem while solving it.

4 Sources, 35d ago, first seen 35d ago

Featured in

EasyPPO is pitched as more stable PPO for LLM post-training

Human coders reportedly show super-linear scaling on 14-day tasks

Featured in

EasyPPO is pitched as more stable PPO for LLM post-training

Human coders reportedly show super-linear scaling on 14-day tasks