TRUST computes confidence as the angular distance between a test image and a slightly modified, maximally-confident version of it, and claims this ranks predictions monotonically.
AI in Health: Keeping the Human in the Loop, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TRUST: Test-time Resource Utilization for Superior Trustworthiness
TRUST computes confidence as the angular distance between a test image and a slightly modified, maximally-confident version of it, and claims this ranks predictions monotonically.