mlsanityData to 14 Oct 2026

Interpretable Uncertainty Routing Separating Emotion Ambiguity from Distribution Shift in Facial Expression Recognition

arXiv 2606.22725 ↗First posted 21 Jun 2026cs.CV cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Facial expression recognition (FER) is inherently ambiguous: human annotators frequently disagree, and models deployed in real environments face distribution shift. Crucially, these two conditions demand different downstream actions, as ambiguous in-distribution faces should be reported with their ambiguity whereas out-of-distribution inputs should be rejected. However, a single uncertainty score conflates the two. In this study, uncertainty decomposition into aleatoric and epistemic components for FER is investigated, and Uncertainty-Aware Routing (UAR), an inference-time routing mechanism that exploits the separation, is introduced. Specifically, aleatoric and epistemic uncertainties are obtained from a Deep Ensemble of fully fine-tuned DINOv2 models and are each validated against an independent external signal: aleatoric against human annotator disagreement, and epistemic against distribution shift induced by image corruptions. The proposed dual-validation protocol reveals that aleatoric recovers annotator disagreement with Spearman correlation 0.66 (95% CI: 0.64-0.68), and epistemic detects corruption-induced shifts, achieving average AUROC of 0.699 at the highest corruption severity. UAR retains approximately 1.8 times more ambiguous in-distribution faces than single-uncertainty routing at a matched out-of-distribution rejection rate. A strong label-distribution-learning baseline achieves comparable disagreement recovery but cannot separate ambiguity from shift and therefore cannot route, establishing that the value of decomposition lies in the separation enabling interpretable and differentiated action selection.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 24 of its 34 references that are in the map. First posted on arXiv 21 Jun 2026.

Builds on

24 of 34 references.

References · 24

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  3. 2017On Calibration of Modern Neural Networkscited by 4,960
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  5. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  6. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  7. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  8. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  9. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,287
  10. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 913
  11. 2017Selective Classification for Deep Neural Networkscited by 852
  12. 2019Addressing Failure Prediction by Learning Model Confidencecited by 186
12 more
  1. 2016Training Deep Networks for Facial Expression Recognition with Crowd-Sourced Label Distributioncited by 158
  2. 2020Suppressing Uncertainties for Large-Scale Facial Expression Recognitioncited by 157
  3. 2022The 'Problem' of Human Label Variation: On Ground Truth in Data, Modeling and Evaluationcited by 139
  4. 2014Label Distribution Learningcited by 124
  5. 2021Dive into Ambiguity: Latent Distribution Mining and Pairwise Uncertainty Estimation for Facial Expression Recognitioncited by 89
  6. 2022Learn From All: Erasing Attention Consistency for Noisy Label Facial Expression Recognitioncited by 67
  7. 2022Uncertainty-aware Label Distribution Learning for Facial Expression Recognitioncited by 30
  8. 2023LA-Net: Landmark-Aware Learning for Reliable Facial Expression Recognition under Label Noisecited by 17
  9. 2026LLMs Capture Emotion Labels, Not Emotion Uncertainty: Distributional Analysis and Calibration of Human-LLM Judgment Gapscited by 4
  10. 2025Navigating Label Ambiguity for Facial Expression Recognition in the Wildcited by 2
  11. 2026Uncertainty Decomposition via Cyclical SG-MCMC and Soft-label Learning for Subjective NLPcited by 2
  12. 2026Bayesian Spectral Emotion Transition Discovery from Multi-Annotator Disagreementcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Who Determines the Meaning of an Emotion? Affective Sovereignty as an Epistemic Consequence of Measurement Limitscited by 0
  2. 2026Predicting Human Disagreement for Calibrated Dynamic Facial Expression Recognitioncited by 0