Author

Rui Wu

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Jun 2026

Greed Is Learned: Visible Incentives as Reward-Hacking Triggers

In MoneyWorld, a synthetic workplace environment, five instruction-tuned models from three families with RL are trained on non-safety tasks in which a visible payoff signal identifies a rewarded shortcut that sacrifices task quality.

Tong Che, Rui Wu · arXiv.org · 0 citations