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

Hidden objective detection research benchmarks · SOTA2 Research

Hidden objective detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Prompt-Manipulation Out-of-Domain Evaluation Stealth Strategy

Gemini-1.5-Flash

100Exit Rate

44

May 14, 2026

Hidden Objective Detection Evaluation Set Dependency 1.0 (out-of-domain)

100Exit Rate

44

May 14, 2026

Hidden Objective Detection Evaluation Set Monitor Notice 1.0 (out-of-domain)

100Exit

22

May 14, 2026

9 objectives out-of-domain code-manipulation A2

96.67Exit Rate

7

May 14, 2026

100Evasion Success Rate

6

Mar 24, 2026

Stealth task trajectories (pooled across american_airlines, day_trip, extract_email, and send_link tasks)

98Hidden Objective 1 Score

3

Mar 24, 2026

Popular datasets

Stealth task trajectories send-link 9 objectives Hidden Objective Detection Evaluation Set Hidden Objective Detection

Follow for updates

Join SOTA2 Contact