# VersaVogue: Visual Expert Orchestration and Preference Alignment for Unified Fashion Synthesis

arXiv: https://arxiv.org/abs/2604.07210 · HTML: https://mlsanity.com/paper/2604.07210/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models.
- Builds on 21 of its 46 references that are in the map.
- First posted on arXiv 8 Apr 2026.

## Placed in

- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)
- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference](https://mlsanity.com/paper/2310.04378/) (2023)
- [BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing](https://mlsanity.com/paper/2305.14720/) (2023)
- [VITON-HD: High-Resolution Virtual Try-On via Misalignment-Aware Normalization](https://mlsanity.com/paper/2103.16874/) (2021)
- [CogVLM2: Visual Language Models for Image and Video Understanding](https://mlsanity.com/paper/2408.16500/) (2024)
- [StableVITON: Learning Semantic Correspondence with Latent Diffusion Model for Virtual Try-On](https://mlsanity.com/paper/2312.01725/) (2023)
- [Dress Code: High-Resolution Multi-Category Virtual Try-On](https://mlsanity.com/paper/2204.08532/) (2022)
- [OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on](https://mlsanity.com/paper/2403.01779/) (2024)
- [Scaling Diffusion Transformers to 16 Billion Parameters](https://mlsanity.com/paper/2407.11633/) (2024)
- [StableGarment: Garment-Centric Generation via Stable Diffusion](https://mlsanity.com/paper/2403.10783/) (2024)
- [Magic Clothing: Controllable Garment-Driven Image Synthesis](https://mlsanity.com/paper/2404.09512/) (2024)
- [DiffCloth: Diffusion Based Garment Synthesis and Manipulation via Structural Cross-modal Semantic Alignment](https://mlsanity.com/paper/2308.11206/) (2023)
- [FastFit: Accelerating Multi-Reference Virtual Try-On via Cacheable Diffusion Models](https://mlsanity.com/paper/2508.20586/) (2025)
- [RL-LOGO: Deep Reinforcement Learning Localization for Logo Recognition](https://mlsanity.com/paper/2312.16792/) (2023)
- [AnyLogo: Symbiotic Subject-Driven Diffusion System with Gemini Status](https://mlsanity.com/paper/2409.17740/) (2024)

## Abstract (arXiv)

Diffusion models have driven remarkable advancements in fashion image generation, yet prior works usually treat garment generation and virtual dressing as separate problems, limiting their flexibility in real-world fashion workflows. Moreover, fashion image synthesis under multi-source heterogeneous conditions remains challenging, as existing methods typically rely on simple feature concatenation or static layer-wise injection, which often causes attribute entanglement and semantic interference. To address these issues, we propose VersaVogue, a unified framework for multi-condition controllable fashion synthesis that jointly supports garment generation and virtual dressing, corresponding to the design and showcase stages of the fashion lifecycle. Specifically, we introduce a trait-routing attention (TA) module that leverages a mixture-of-experts mechanism to dynamically route condition features to the most compatible experts and generative layers, enabling disentangled injection of visual attributes such as texture, shape, and color. To further improve realism and controllability, we develop an automated multi-perspective preference optimization (MPO) pipeline that constructs preference data without human annotation or task-specific reward models. By combining evaluators of content fidelity, textual alignment, and perceptual quality, MPO identifies reliable preference pairs, which are then used to optimize the model via direct preference optimization (DPO). Extensive experiments on both garment generation and virtual dressing benchmarks demonstrate that VersaVogue consistently outperforms existing methods in visual fidelity, semantic consistency, and fine-grained controllability.