EUVIMEDEUVIMED entdecken
Crossref

Alfa: Attentive Low-Rank Filter Adaptation for Structure-Aware Cross-Domain Personalized Gaze Estimation

He-Yen Hsieh, Wei-Te Mark Ting, H. T. Kung

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (21) · 17481-17489

DOI: 10.1609/aaai.v40i21.38802 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von He-Yen Hsieh und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pre-trained gaze models learn to identify useful patterns commonly found across users, but subtle user-specific variations (i.e., eyelid shape or facial structure) can degrade model performance. Test-time personalization (TTP) adapts pre-trained models to these user-specific domain shifts using only a few unlabeled samples. Efficient fine-tuning is critical in performing this domain adaptation: data and computation resources can be limited-especially for on-device customization. While popular parameter-efficient fine-tuning (PEFT) methods address adaptation costs by updating only a small set of weights, they may not be taking full advantage of structures encoded in pre-trained filters. To more effectively leverage existing structures learned during pre-training, we reframe personalization as a process to reweight existing features rather than learning entirely new ones. We present Attentive Low-Rank Filter Adaptation (Alfa) to adapt gaze models by reweighting semantic patterns in pre-trained filters. With Alfa, singular value decomposition (SVD) extracts dominant spatial components that capture eye and facial characteristics across users. Via an attention mechanism, we need only a few unlabeled samples to adjust and reweight pre-trained structures, selectively amplifying those relevant to a target user. Alfa achieves the lowest average gaze errors across four cross-dataset gaze benchmarks, outperforming existing TTP methods and low-rank adaptation (LoRA)-based variants. We also show that Alfa's attentive low-rank methods can be applied to applications beyond vision, such as diffusion-based language models.

Zitieren

APA

Hsieh, H., Ting, W. M., & Kung, H. T. (2026). Alfa: Attentive Low-Rank Filter Adaptation for Structure-Aware Cross-Domain Personalized Gaze Estimation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(21), 17481-17489. https://doi.org/10.1609/aaai.v40i21.38802

Vancouver

Hsieh H, Ting WM, Kung HT. Alfa: Attentive Low-Rank Filter Adaptation for Structure-Aware Cross-Domain Personalized Gaze Estimation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(21):17481-17489. doi:10.1609/aaai.v40i21.38802

BibTeX
@article{Hsieh2026,
  author = {Hsieh, He-Yen and Ting, Wei-Te Mark and Kung, H. T.},
  title = {Alfa: Attentive Low-Rank Filter Adaptation for Structure-Aware Cross-Domain Personalized Gaze Estimation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {21},
  pages = {17481--17489},
  doi = {10.1609/aaai.v40i21.38802}
}
Kostenlos starten →