Metraipolis › archive › dc_507da540dfa3

Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond — arXiv:2609.39573

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39573
License (as archived)
CC0-1.0
Archived
2026-10-07T12:45:31.836Z
SHA-256 of the archived text
b43de8f4dacae963215c2d2e91f5c3f649e27d3a8e70fcadc90bdcc60858d8f5
Stable URL of this page
https://metraipolis.com/documento/dc_507da540dfa3

How to cite

Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond — arXiv:2609.39573 — arXiv (en), https://arxiv.org/abs/2609.39573 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_507da540dfa3 (Metraipolis archive dc_507da540dfa3, sha256 b43de8f4dacae963215c2d2e91f5c3f649e27d3a8e70fcadc90bdcc60858d8f5)
BibTeX
@misc{dc507da540dfa3,
  title = {Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond — arXiv:2609.39573},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39573},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_507da540dfa3; sha256 b43de8f4dacae963215c2d2e91f5c3f649e27d3a8e70fcadc90bdcc60858d8f5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39573. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As state-of-the-art text-to-image flow models achieve near-photorealistic quality, controlling their outputs, e.g., suppressing harmful content while promoting benign alternatives, has become a central challenge. The current steering paradigm consists of adding a global steering vector to selected activations. While functional, a fixed and example-agnostic vector applied uniformly along the entire trajectory cannot adapt to the changing state of the generation and often causes unintended global changes. We introduce Steering Fields, a generalization of steering vectors that adaptively re-estimates the steering direction at each step of the generative process. Steering Fields operate on the noisy states of flow models, expose a continuous trade-off between steering strength and content preservation, and are compositional, enabling the simultaneous induction and inhibition of concepts, setting a new state of the art on safety steering benchmarks. Despite using no explicit spatial masks or object priors, the trajectory-adaptive estimation naturally preserves local structure, in a manner reminiscent of image editing. In fact, Steering Fields can serve as a structure-preserving image-editing technique that achieves state-of-the-art semantic fidelity (CLIP, VQAScore), while remaining model-agnostic and inversion-free.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.39573 — 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 2026-10-07T12:14:14.107Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).