# OPRO: Orthogonal Panel-Relative Operators for Panel-Aware In-Context Image Generation Source: [OPRO: Orthogonal Panel-Relative Operators for Panel-Aware In-Context Image Generation]() Collection index: [llms.txt]() Data updated: 2026-04-01T05:00:12.613Z Authors: Sanghyeon Lee, Minwoo Lee, Euijin Shin, Kangyeol Kim, Seunghwan Choi, Jaegul Choo Published: 2026-03-29T11:21:19.000Z ## Abstract We introduce a parameter-efficient adaptation method for panel-aware in-context image generation with pre-trained diffusion transformers. The key idea is to compose learnable, panel-specific orthogonal operators onto the backbone's frozen positional encodings. This design provides two desirable properties: (1) isometry, which preserves the geometry of internal features, and (2) same-panel invariance, which maintains the model's pre-trained intra-panel synthesis behavior. Through controlled experiments, we demonstrate that the effectiveness of our adaptation method is not tied to a specific positional encoding design but generalizes across diverse positional encoding regimes. By enabling effective panel-relative conditioning, the proposed method consistently improves in-context image-based instructional editing pipelines, including state-of-the-art approaches. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Instructive Image Editing on MagicBrush (test)](): Instructive image editing; MagicBrush (test); CLIP Image: 0.9281. - [Subject-driven image generation on DreamBooth Dataset 1.0 (test)](): Subject-driven image generation; DreamBooth Dataset 1.0 (test); DINO Score: 0.6192. ## Original sources and implementation [Original paper]() [PDF]()