SOURCE-LINKED INTELLIGENCE

Predicting Human Disagreement for Calibrated Dynamic Facial Expression Recognition

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Dynamic facial expression recognition (DFER) benchmarks such as DFEW provide multiple annotator votes per clip, yet most models collapse them to a majority label and cannot represent human disagreement at inference time. We propose a disagreement-aware DFER framework that trains directly on the raw annotator count vector using a Dirichlet-Multinomial likelihood. Unlike mean-only soft-label objectives, the proposed likelihood provides scale-sensitive supervision for the Dirichlet concentration while preserving the predictive mean. A separate ambiguity head predicts annotation entropy for unseen

Read original source ↗ Open in workspace

recordType

paper

region

Global