Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Reward Hacking Detection research benchmarks · SOTA2 Research

Reward Hacking Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

School of Reward Hacks and Stanford Alpaca (test)

activation-based monitoring system

100F1 Score

18

Mar 5, 2026

Anthropic’s Auditing Game

96.6Acc (Poisoned)

6

Feb 26, 2026

Popular datasets

Anthropic’s Auditing Game School of Reward Hacks and Stanford Alpaca

Follow for updates

Join SOTA2 Contact