ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911530992271360 |
|---|---|
| author | Sepehri, Mohammad Shahab Mehradfar, Asal Tinaz, Berk Avestimehr, Salman Soltanolkotabi, Mahdi |
| author_facet | Sepehri, Mohammad Shahab Mehradfar, Asal Tinaz, Berk Avestimehr, Salman Soltanolkotabi, Mahdi |
| contents | Text-to-image diffusion models achieve high visual fidelity but surprisingly exhibit systematic failures in numerical control when prompts specify explicit object counts. To address this limitation, we introduce ATHENA, a model-agnostic, test-time adaptive steering framework that improves object count fidelity without modifying model architectures or requiring retraining. ATHENA leverages intermediate representations during sampling to estimate object counts and applies count-aware noise corrections early in the denoising process, steering the generation trajectory before structural errors become difficult to revise. We present three progressively more advanced variants of ATHENA that trade additional computation for improved numerical accuracy, ranging from static prompt-based steering to dynamically adjusted count-aware control. Experiments on established benchmarks and a new visually and semantically complex dataset show that ATHENA consistently improves count fidelity, particularly at higher target counts, while maintaining favorable accuracy-runtime trade-offs across multiple diffusion backbones. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_19676 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models Sepehri, Mohammad Shahab Mehradfar, Asal Tinaz, Berk Avestimehr, Salman Soltanolkotabi, Mahdi Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Text-to-image diffusion models achieve high visual fidelity but surprisingly exhibit systematic failures in numerical control when prompts specify explicit object counts. To address this limitation, we introduce ATHENA, a model-agnostic, test-time adaptive steering framework that improves object count fidelity without modifying model architectures or requiring retraining. ATHENA leverages intermediate representations during sampling to estimate object counts and applies count-aware noise corrections early in the denoising process, steering the generation trajectory before structural errors become difficult to revise. We present three progressively more advanced variants of ATHENA that trade additional computation for improved numerical accuracy, ranging from static prompt-based steering to dynamically adjusted count-aware control. Experiments on established benchmarks and a new visually and semantically complex dataset show that ATHENA consistently improves count fidelity, particularly at higher target counts, while maintaining favorable accuracy-runtime trade-offs across multiple diffusion backbones. |
| title | ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2603.19676 |