Mitigating Diffusion Model Hallucinations with Dynamic Guidance
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908579076767744 |
|---|---|
| author | Triaridis, Kostas Graikos, Alexandros Chatziagapi, Aggelina Chrysos, Grigorios G. Samaras, Dimitris |
| author_facet | Triaridis, Kostas Graikos, Alexandros Chatziagapi, Aggelina Chrysos, Grigorios G. Samaras, Dimitris |
| contents | Diffusion models, despite their impressive demos, often produce hallucinatory samples with structural inconsistencies that lie outside of the support of the true data distribution. Such hallucinations can be attributed to excessive smoothing between modes of the data distribution. However, semantic interpolations are often desirable and can lead to generation diversity, thus we believe a more nuanced solution is required. In this work, we introduce Dynamic Guidance, which tackles this issue. Dynamic Guidance mitigates hallucinations by selectively sharpening the score function only along the pre-determined directions known to cause artifacts, while preserving valid semantic variations. To our knowledge, this is the first approach that addresses hallucinations at generation time rather than through post-hoc filtering. Dynamic Guidance substantially reduces hallucinations on both controlled and natural image datasets, significantly outperforming baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_05356 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mitigating Diffusion Model Hallucinations with Dynamic Guidance Triaridis, Kostas Graikos, Alexandros Chatziagapi, Aggelina Chrysos, Grigorios G. Samaras, Dimitris Computer Vision and Pattern Recognition Machine Learning Diffusion models, despite their impressive demos, often produce hallucinatory samples with structural inconsistencies that lie outside of the support of the true data distribution. Such hallucinations can be attributed to excessive smoothing between modes of the data distribution. However, semantic interpolations are often desirable and can lead to generation diversity, thus we believe a more nuanced solution is required. In this work, we introduce Dynamic Guidance, which tackles this issue. Dynamic Guidance mitigates hallucinations by selectively sharpening the score function only along the pre-determined directions known to cause artifacts, while preserving valid semantic variations. To our knowledge, this is the first approach that addresses hallucinations at generation time rather than through post-hoc filtering. Dynamic Guidance substantially reduces hallucinations on both controlled and natural image datasets, significantly outperforming baselines. |
| title | Mitigating Diffusion Model Hallucinations with Dynamic Guidance |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2510.05356 |