Dodgersort: Uncertainty-Aware VLM-Guided Human-in-the-Loop Pairwise Ranking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Park, Yujin, Chung, Haejun, Jang, Ikbeom
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911534465155072
author Park, Yujin
Chung, Haejun
Jang, Ikbeom
author_facet Park, Yujin
Chung, Haejun
Jang, Ikbeom
contents Pairwise comparison labeling is emerging as it yields higher inter-rater reliability than conventional classification labeling, but exhaustive comparisons require quadratic cost. We propose Dodgersort, which leverages CLIP-based hierarchical pre-ordering, a neural ranking head and probabilistic ensemble (Elo, BTL, GP), epistemic--aleatoric uncertainty decomposition, and information-theoretic pair selection. It reduces human comparisons while improving the reliability of the rankings. In visual ranking tasks in medical imaging, historical dating, and aesthetics, Dodgersort achieves a 11--16\% annotation reduction while improving inter-rater reliability. Cross-domain ablations across four datasets show that neural adaptation and ensemble uncertainty are key to this gain. In FG-NET with ground-truth ages, the framework extracts 5--20$\times$ more ranking information per comparison than baselines, yielding Pareto-optimal accuracy--efficiency trade-offs.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20839
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Dodgersort: Uncertainty-Aware VLM-Guided Human-in-the-Loop Pairwise Ranking
Park, Yujin
Chung, Haejun
Jang, Ikbeom
Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
Machine Learning
Pairwise comparison labeling is emerging as it yields higher inter-rater reliability than conventional classification labeling, but exhaustive comparisons require quadratic cost. We propose Dodgersort, which leverages CLIP-based hierarchical pre-ordering, a neural ranking head and probabilistic ensemble (Elo, BTL, GP), epistemic--aleatoric uncertainty decomposition, and information-theoretic pair selection. It reduces human comparisons while improving the reliability of the rankings. In visual ranking tasks in medical imaging, historical dating, and aesthetics, Dodgersort achieves a 11--16\% annotation reduction while improving inter-rater reliability. Cross-domain ablations across four datasets show that neural adaptation and ensemble uncertainty are key to this gain. In FG-NET with ground-truth ages, the framework extracts 5--20$\times$ more ranking information per comparison than baselines, yielding Pareto-optimal accuracy--efficiency trade-offs.
title Dodgersort: Uncertainty-Aware VLM-Guided Human-in-the-Loop Pairwise Ranking
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2603.20839