# Learning to Solve Hard Problems in RL for LLMs by Never Giving Up

> Source: <https://aiflash.com/news/120291/>
> Published: 2026-09-15 19:30:27+00:00

We demonstrate that training LLMs with RL does not improve performance equally across a dataset. RL shows large improvements on easy problems that an LLM is already good at solving, but small improvements on hard problems. We call this the Matthew Effect in RL for LLMs, after the phenomenon of cumul