Qwen reward-hacked models collapse into random tokens — with unexpectedly more swearing

voooooogel · x · 2026-09-01

Developer voooooogel shares a curious observation: Qwen models that reward-hacked during his experiments would collapse into random tokens, yet those random tokens contained far more swearing and crudeness than expected. He notes this kind of collapse data is genuinely interesting and wishes labs published more of it.

Original post →

More from Fun

Fun channel →