Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Subject and Property Perturbations

Benchmarks

Task NameDataset NameSOTA ResultTrend
Output ConsistencySubject and Property Perturbations
Subject Agreement91
24
Input-bound consistencySubject and Property Perturbations dataset
Subject Agreement (%)38
14
Showing 2 of 2 rows