cs.CVNov 17, 2026

Breaking Visual Conventions: Spectrally Coupled Weight Perturbation for Creative Image Generation

Authors: Fu Feng, Ruoheng Zhou, Yucheng Xie, Xu Yang, Jing Wang, Xin Geng

Organizations: School of Computer Science and Engineering, Southeast University, Nanjing, China · Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, China

Abstract

Text-to-image (T2I) models often produce high-quality images that follow familiar visual conventions, even in response to open-ended prompts such as ``a creative [CLS]''. Existing methods encourage novelty by exploring semantic conditions or modifying features and guidance during denoising. Pretrained weights offer a complementary source of creative variation, but unstructured perturbations can have little visual effect or degrade image quality. We introduce Coupled Spectral Perturbation (CosPert), a training-free method that targets attention projections in layers identified by a one-time sensitivity analysis. For each projection, CosPert uses a shared random transformation and its transpose to couple updates to the singular-vector factors while perturbing the spectral gains. A common relative budget calibrates the magnitude of weight changes across projections, and a sigmoid schedule concentrates their influence in early denoising and gradually attenuates them during refinement. CosPert provides plug-and-play creativity enhancement at inference time. Once the perturbations are prepared, sampling requires no auxiliary networks or additional model evaluations. On both FLUX.1 and Stable Diffusion 3.5, CosPert enhances visual creativity, producing more varied object structures while preserving category identity. Across 50 object categories, CosPert achieves the highest mean CREward among the evaluated methods while maintaining strong prompt alignment.

Figures & tables

Explore similar work

CardsList
  1. Self-Creative Text-to-Object Generation using Semantic-Aware Spatial Weighting

    May 19, 2026Yue Yu, Haibo Chen, Shuo Chen +2T2I GenerationComputational Creativity

  2. Concept Guidance: Precise, Training-Free Latent Control for Text-to-Image Generation

    Aug 14, 2026Nikolai Röhrich, Isabell Hans, Felix Krause +1Diffusion Model GuidanceT2I Generation

  3. A Creative Agent is Worth a 64-Token Template

    Mar 18, 2026Ruixiao Shi, Fu Feng, Yucheng Xie +3T2I GenerationImage Generation