Evidence · DiggingBeagle record
ObligationBench shows omissions can make agents unsafe despite no forbidden action
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-096. Benchmark finding, not universal omission prevalence.
- Source role
- primary disclosure
Evidence record
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-096. Benchmark finding, not universal omission prevalence.
Claim-level citations (3)
- supportsObligationBench highlights security failures caused by skipped agent checks: In the paperâs preliminary GLM-5.3 sample, 56.92% of trajectories contained an unfulfilled obligation and 30% a forbidden action.
Abstract, preliminary trajectory study
- supportsObligationBench highlights security failures caused by skipped agent checks: ObligationBench contains 240 expert-validated trajectories; the authors report ObligationGuard recall of 57.52%, compared with a best 48.97% among tested baselines.
Abstract, benchmark construction and model evaluations
- supportsObligationBench highlights security failures caused by skipped agent checks: The evaluated 14 baseline guard models reached at most 10% exact match; the authors report 21.67% exact match for ObligationGuard trained on 40,000 synthetic examples.
Abstract, baseline model comparison and ObligationGuard results