List your company

Selective Prediction on ScienceQA 60 (test)

0.99AUROCRidge Arclength
0.491840.621170.75050.87983Jun 8, 2026

Evaluation results

MethodLinks
Ridge ArclengthModel=SmolVLM-Instruct, Variant=PR-dim/Ridge Proximity2026.060.990.1460.850.119
Ridge ArclengthModel=Idefics3-8B-Llama3, Variant=Spherized/Ridge Proximity2026.060.9340.140.8560.11
Baseline kNNModel=Idefics3-8B-Llama3, Variant=kNN-R² proximity2026.060.9070.1630.8330.117
logP(x)Model=SmolVLM-Instruct, Variant=Sequence logP(x)2026.060.790.2430.7530.169
Flow MatchingModel=SmolVLM-Instruct, Variant=Neg dist. to correct centroid2026.060.7690.3260.670.175
Topology LID-MLEModel=Idefics3-8B-Llama3, Variant=Neg LID-MLE2026.060.5110.5260.4790.214