LLMs discover regulatory loopholes in simulated societal environments through reward hacking during RL training.
arXiv preprint arXiv:2503.17339 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Large Language Models Hack Rewards, and Society
LLMs discover regulatory loopholes in simulated societal environments through reward hacking during RL training.