AffectFlow-DINO: Uncertainty-Aware Multi-Task Affect Estimation via Conditional Rectified Flow
:::info Stub — Full Engineering Breakdown Coming This paper was featured on Hugging Face Daily Papers on 2026-07-14 with 4 upvotes. A full breakdown with production viability rating, implementation notes, and honest limitations is being written. Subscribe to AI Letters → :::
| Authors | Salah Eddine Bekhouche et al. |
| Year | 2026 |
| HF Upvotes | 4 |
| arXiv | 2607.13250 |
| Download | |
| HF Page | View on Hugging Face |
Abstract
We present AffectFlow-DINO, a multi-task learning system for the 11th ABAW challenge that extends a standard deterministic architecture with a conditional rectified-flow head to model the inherent ambiguity of in-the-wild facial behavior. Instead of predicting a single affect estimate, the model learns a conditional generative distribution, enabling uncertainty-aware one-to-many predictions through Monte Carlo sampling. The system jointly estimates continuous valence-arousal, classifies eight facial expressions, and detects twelve Action Units from static face images. Built on a frozen DINOv3 ViT-S/16 backbone, extensive ablation studies show that rectified-flow decoding consistently improves deterministic prediction, particularly for valence-arousal estimation (CCC-V +0.058). We further show that post-hoc threshold calibration effectively recovers performance on severely imbalanced rare classes (e.g., Fear: 3.8% rightarrow 33.1%) without retraining. Combined with backbone fine-tuning and flow retuning, the final model achieves P_{MTL=1.177}, substantially outperforming the official challenge baseline of P_{MTL}=0.45.
Engineering Breakdown
The Problem
Combined with backbone fine-tuning and flow retuning, the final model achieves P_{MTL=1.177}, substantially outperforming the official challenge baseline of P_{MTL}=0.45.
The Approach
We present AffectFlow-DINO, a multi-task learning system for the 11th ABAW challenge that extends a standard deterministic architecture with a conditional rectified-flow head to model the inherent ambiguity of in-the-wild facial behavior.
Key Results
Combined with backbone fine-tuning and flow retuning, the final model achieves P_{MTL=1.177}, substantially outperforming the official challenge baseline of P_{MTL}=0.45.
Research Areas
This paper contributes to the following areas of AI/ML engineering:
- Machine learning
- Deep learning
- Neural networks
- Model optimization
- AI systems
- Affectflowdino
:::tip Subscribe Get weekly breakdowns of papers like this in AI Letters - the newsletter for engineers building production AI systems. :::