# AutoThreshold: Joint Confusion-Profile Learning for Adaptive Mask Binarization

**Authors:** Shaokai Wu, Xunchu Zhou, Pinzhi Wang, Bo Dong, Xiao Cui, Qi Kong, Yue Ding, Hongtao Lu  
**Published:** 10 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.14279) · [PDF](https://arxiv.org/pdf/2611.14279) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.14279.html)

## Abstract

A single validation-tuned threshold can become brittle when dense-score distributions shift. We introduce AutoThreshold, a post-hoc policy that predicts the joint mass of positive and negative pixels over score bins. This target directly parameterizes threshold-indexed confusion counts, avoiding products of independently estimated prevalence and class-conditional rates. Softmax normalization and cumulative decoding yield coherent, monotone confusion profiles; a plug-in decoder then selects a threshold for any F_β objective without retraining the head. Across three independently trained Kvasir pipelines transferred to four polyp datasets, Joint exceeds continuous Otsu by 0.0080 ± 0.0046 external macro Dice, with positive gains in 3/3 runs. On the reference pipeline, three head seeds achieve 0.6324 ± 0.0027, 0.0085 above Otsu; direct-utility Ridge achieves 0.6342, with the paired difference interval spanning zero. The same Joint checkpoints outperform Otsu for F_0.5, Dice, and F_2. On developmental KITTI Road leave-one-corruption-family-out evaluation, Joint improves over scalar regression by 0.0082 [0.0064, 0.0101], while direct utility is 0.0022 higher. These results establish joint confusion prediction as a useful, metric-retargetable inductive bias, without claiming Bayes-optimal F_β decisions.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.