Metraipolis › archive › dc_cc3af01251b2

TACO: Trajectory-Aware Curvature Observation for Memorization Localization and Mitigation in Diffusion Models — arXiv:2610.41520

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41520
License (as archived)
CC0-1.0
Archived
2026-11-03T05:46:25.774Z
SHA-256 of the archived text
f970b6d077fd1441c763ee477d25fdb985d6951cf000674a6108c97c7a88e868
Stable URL of this page
https://metraipolis.com/documento/dc_cc3af01251b2

How to cite

TACO: Trajectory-Aware Curvature Observation for Memorization Localization and Mitigation in Diffusion Models — arXiv:2610.41520 — arXiv (en), https://arxiv.org/abs/2610.41520 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cc3af01251b2 (Metraipolis archive dc_cc3af01251b2, sha256 f970b6d077fd1441c763ee477d25fdb985d6951cf000674a6108c97c7a88e868)
BibTeX
@misc{dccc3af01251b2,
  title = {TACO: Trajectory-Aware Curvature Observation for Memorization Localization and Mitigation in Diffusion Models — arXiv:2610.41520},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41520},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cc3af01251b2; sha256 f970b6d077fd1441c763ee477d25fdb985d6951cf000674a6108c97c7a88e868},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41520. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Text-to-image diffusion models can reproduce memorized training content within localized image regions, making precise localization essential for targeted mitigation. However, curvature observations at a single denoising state can contain transient responses, while spatially uniform mitigation may disrupt prompt-relevant content. We present Trajectory-Aware Curvature Observation (TACO), a unified framework for local memorization detection and mitigation in diffusion models. Our empirical analysis reveals that memorization-related curvature becomes pronounced during late-stage denoising and persists at consistent spatial locations across neighboring states. Motivated by these observations, TACO aggregates prompt-dependent curvature responses along the late denoising trajectory and applies latent-guided spatial refinement to improve spatial coherence. The resulting memorization map serves both as a localization signal and as a spatial control signal for mitigation. Specifically, TACO uses this map to steer existing inference-time interventions toward memorization-related regions through spatial reweighting and localized guidance correction. Extensive experiments demonstrate that TACO consistently outperforms competing localization methods, and improves the Pareto trade-off between memorization suppression, semantic alignment, and perceptual quality. These results establish trajectory-aware spatial analysis as an effective link between memorization localization and targeted mitigation.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.41520 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T05:33:32.302Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).