FAIRT2V: Training-Free Debiasing for Text-to-Video Diffusion Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908794298040320 |
|---|---|
| author | Zhong, Haonan Song, Wei Han, Tingxu Pagnucco, Maurice Xue, Jingling Song, Yang |
| author_facet | Zhong, Haonan Song, Wei Han, Tingxu Pagnucco, Maurice Xue, Jingling Song, Yang |
| contents | Text-to-video (T2V) diffusion models have achieved rapid progress, yet their demographic biases, particularly gender bias, remain largely unexplored. We present FairT2V, a training-free debiasing framework for text-to-video generation that mitigates encoder-induced bias without finetuning. We first analyze demographic bias in T2V models and show that it primarily originates from pretrained text encoders, which encode implicit gender associations even for neutral prompts. We quantify this effect with a gender-leaning score that correlates with bias in generated videos.
Based on this insight, FairT2V mitigates demographic bias by neutralizing prompt embeddings via anchor-based spherical geodesic transformations while preserving semantics. To maintain temporal coherence, we apply debiasing only during early identity-forming steps through a dynamic denoising schedule. We further propose a video-level fairness evaluation protocol combining VideoLLM-based reasoning with human verification. Experiments on the modern T2V model Open-Sora show that FairT2V substantially reduces demographic bias across occupations with minimal impact on video quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20791 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FAIRT2V: Training-Free Debiasing for Text-to-Video Diffusion Models Zhong, Haonan Song, Wei Han, Tingxu Pagnucco, Maurice Xue, Jingling Song, Yang Computer Vision and Pattern Recognition Artificial Intelligence Text-to-video (T2V) diffusion models have achieved rapid progress, yet their demographic biases, particularly gender bias, remain largely unexplored. We present FairT2V, a training-free debiasing framework for text-to-video generation that mitigates encoder-induced bias without finetuning. We first analyze demographic bias in T2V models and show that it primarily originates from pretrained text encoders, which encode implicit gender associations even for neutral prompts. We quantify this effect with a gender-leaning score that correlates with bias in generated videos. Based on this insight, FairT2V mitigates demographic bias by neutralizing prompt embeddings via anchor-based spherical geodesic transformations while preserving semantics. To maintain temporal coherence, we apply debiasing only during early identity-forming steps through a dynamic denoising schedule. We further propose a video-level fairness evaluation protocol combining VideoLLM-based reasoning with human verification. Experiments on the modern T2V model Open-Sora show that FairT2V substantially reduces demographic bias across occupations with minimal impact on video quality. |
| title | FAIRT2V: Training-Free Debiasing for Text-to-Video Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2601.20791 |