pith. sign in

Recoverable Identifier

arXiv:2604.26348 · detector doi_compliance · incontrovertible · 2026-05-19 20:13:41.322581+00:00

advisory doi_compliance recoverable_identifier

DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1109/cvpr.2018.58ACPO) was visible in the surrounding text but could not be confirmed against doi.org as printed.

Paper page Integrity report arXiv Try DOI

Evidence text

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. InIEEE/CVF Conference on Computer Vision and Pattern Recognition. 586–595. doi:10.1109/CVPR.2018.58 ACPO: Anchor-Constrained Perceptual Optimization Yang et al. Supplementary Material This supplementary material complements the main paper by pro- viding an expanded set of visual evaluations and supplementary experimental analyses. Section 1 presents extended qualitative com- parisons for our baseline experiments, offering further insights into the perceptual strengths of our approach. Section 2 showcases a broader range of visual examples illustrating the method’s gen- eralization capabilities across different datasets and diverse text prompts. Finally, Section 3 presents extended experimental anal- yses, including more comprehensive ablation studies conducted on the Stable Diffusion (SD) architecture to evaluate the specific impacts of timestep intervention, quality weighting, and anchor loss. A Extended Qualitative Results This section provides additional qualitative evidence to support the effectiveness of the Anchor-Constrained Perceptual Optimization (ACPO) framework. Following the logic established in the main paper, we showcase results across two dimensions: general percep- tual quality enhancement on unconditional models and semantic alignment refinement on text-to-image models. A.1 Additional Visual Evidence for Perceptual Quality Figure 5 presents a comparative visualization between the base- li

Evidence payload

{
  "printed_excerpt": "The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. InIEEE/CVF Conference on Computer Vision and Pattern Recognition. 586\u2013595. doi:10.1109/CVPR.2018.58 ACPO: Anchor-Constrained Perceptual Optimization Yang et al. Supplem",
  "reconstructed_doi": "10.1109/cvpr.2018.58ACPO",
  "ref_index": 47,
  "resolved_title": null,
  "verdict_class": "incontrovertible"
}