Generalization over Memorization: Generalization-Aware Diffusion Adaptation for Single-Image Multi-View Synthesis

2026-08-29

AI Summary

本文解决了单图像多视角合成问题，通过引入GoM框架，结合场景分离验证、曝光匹配选择和定向扩散适应方法，提高了模型的泛化能力。

Abstract

We present the winning solution to the ACM Multimedia 2026 Grand Challenge on Single-Image Guided Multi-Angle Image Synthesis. It ranks first among 293 registered teams; 56 teams obtained at least one scored submission on the public Phase-A leaderboard. With only 40 training scenes, the challenge requires 26 target views from one RGB model and one forward pass per view; it prohibits explicit geometry, external rendering, chained generation, candidate selection, and post-processing. We identify a critical model-selection failure: shared training and validation scenes make memorization appear as transferable view control. We therefore introduce GoM. Short for Generalization over Memorization, the framework combines scene-disjoint validation, exposure-matched selection, and targeted diffusion adaptation. Its synthesis model adapts a 4B rectified-flow DiT using rank-32 LoRA, optimizer restarts, late-checkpoint averaging, and VAE decoder tuning. More than 300 offline experiments and 24 online submissions show that validation design and training-trajectory control can matter as much as architecture scale in small-data generative modeling.

Problem

• Generalization
• Memorization
• Single-Image Multi-View Synthesis
• Small-Data Generative Modeling
• Validation Design

Innovation

• Generalization over Memorization
• scene-disjoint validation
• diffusion adaptation
• rectified-flow DiT
• LoRA