Evidence-Guided Unknown Rejection for High-Confidence Near-Known Unknowns

TL;DR

This paper introduces EGUR-A, a novel method for open-set recognition that reduces high-confidence false acceptances of near-known unknowns by assessing class-specific evidence rather than relying solely on confidence scores.

Contribution

EGUR-A shifts the decision criterion to evidence-based acceptance, effectively reducing false acceptances of near-known unknowns without needing unknown validation data.

Findings

EGUR-A significantly reduces high-confidence false acceptances across multiple datasets.

The method improves open-set recognition by focusing on evidence for known classes.

EGUR-A operates effectively without unknown validation data.

Abstract

Open-set recognition systems face a neglected failure mode: high-confidence near-known unknowns, which lie outside the known label set but are close enough to known classes that a closed-set classifier accepts them with high confidence. We show that this failure is widespread across scalar-threshold methods, including recent post-hoc detectors, and that stronger encoders can amplify rather than remove the risk. We propose EGUR-A, which changes the decision from ``is this sample's score high enough?'' to ``does this predicted known class have sufficient evidence to accept this sample?'' EGUR-A combines class-conditional local acceptance evidence with global residual evidence, and selects their relative weight from known-sample statistics without unknown validation data. Across CUB, FGVC-Aircraft, and ImageNet-hard, EGUR-A substantially reduces high-confidence false known acceptance at matched known-rejection operating points. The result is not a stronger threshold; it is a different question: whether a known class is entitled to accept a sample.