Compositional Image Generation

Overview

The field aims to make image generators reliably follow instructions about what belongs in a scene, where it goes, how objects relate, and which details come from reference images. Approaches include layout- and graph-conditioned diffusion, spatial and occlusion controls, modular adapters and layered synthesis, and agentic planning with iterative generation and editing. Evaluation spans layout-to-image, multi-subject and multi-reference tasks, with benchmarks such as CogCanvas and TRACE-Bench probing identity, attribute binding, relations, and spatial accuracy. A central challenge is generalizing to unfamiliar combinations and crowded scenes: binding and identity errors grow with complexity, while stronger control must preserve image quality and remain efficient.

Latest papers

CardsList

95 papers8 papers in the last four weeks, down 27% on the four weeks before. 0.1% of all new papers.

  1. REPA-G: Test-Time Conditioning with Representation-Aligned Visual Features

    Feb 3, 2026Nicolas Sereyjol-Garros, Ellington Kirby, Victor Letzelter +2Diffusion Model GuidanceConditional Image Generation

  2. Generative Anchored Fields: Endpoint Learning for Generative Modeling

    Nov 27, 2025Deressa Wodajo Deressa, Hannes Mareen, Peter Lambert +1Generative ModelingConditional Generative Modeling

  3. CanvasComposer: Personalized Group Photo Generation via a Multi-Reference Canvas

    Oct 23, 2025Gordon Guocheng Qian, Ruihang Zhang, Tsai-Shien Chen +11Subject-Driven Image GenerationImage Generation

  4. What Drives Compositional Generalization in Visual Generative Models? The Importance of Continuous Training Objectives

    Oct 3, 2025Karim Farid, Rajat Sahay, Yumna Ali Alnaggar +4Compositional Image GenerationGenerative Modeling