Question: Reward Hacking Detection Metrics in RL-Trained Agents
Specification gaming taxonomy and frequency
Needs new evidence1 verified (technical)
0/5 publication-ready
Findings metrics
—
total
—
verified
—
pending
—
rejected
Question: Reward Hacking Detection Metrics in RL-Trained Agents
—
total
—
verified
—
pending
—
rejected