# Toward Calibrated, Fair, and accurate Deepfake Detection

Source: [Toward Calibrated, Fair, and accurate Deepfake Detection](<https://www.sota2.com/research/paper/6a28f33d90966ade919c16fb>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-11T05:00:05.512Z

Authors: Ryan Brown, Chris Russell

Published: 2026-06-03T05:44:29.000Z

## Abstract

Deepfake detectors show large performance gaps across demographic groups. Existing fairness approaches require demographic labels, retraining, or sacrifice accuracy. We introduce Face-Fairness (FF), a plug-and-play framework for bias mitigation. Our primary contribution, Face-Feature Tuning (FFT), is the first demographic label-free fairness method demonstrated for deepfake detection: a lightweight calibrator that performs a logit remapping conditioned on frozen face embeddings. We complement FFT with two variants: FF-Max, which maximizes worst-group accuracy when demographics are available, and FF-Discover, which does the same with embedding-discovered groups. Across in-domain and cross-dataset test settings, FF consistently reduces FPR/TPR gaps and improves minimum group accuracy while maintaining (often improving) overall accuracy. The approach is detector-agnostic, adds negligible runtime overhead, and requires no access to identity attributes.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Deepfake Detection on FaceForensics++ (LOO)](<https://www.sota2.com/research/sota/deepfake-detection-on-faceforensics-plus-plus-loo.md>): Deepfake Detection; FaceForensics++ (LOO); Accuracy@0.5: 63.92.

- [Deepfake Detection on OpenForensics](<https://www.sota2.com/research/sota/deepfake-detection-on-openforensics.md>): Deepfake Detection; OpenForensics; Accuracy@0.5: 97.48.

- [Deepfake Detection on FF++ LOO](<https://www.sota2.com/research/sota/deepfake-detection-on-ff-plus-plus-loo.md>): Deepfake Detection; FF++ LOO; Acc @ 0.5: 63.76.

- [Deepfake detection on OpenForensics (test)](<https://www.sota2.com/research/sota/deepfake-detection-on-openforensics-test.md>): Deepfake Detection; OpenForensics (test); Accuracy: 96.48.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.09881>)

[PDF](<https://arxiv.org/pdf/2606.09881.pdf>)