StyleTailor: Towards Personalized Fashion Styling via Hierarchical Negative Feedback

TL;DR

StyleTailor is a novel collaborative framework that uses hierarchical negative feedback to improve personalized fashion styling, recommendation, and virtual try-on, achieving superior results in a comprehensive evaluation suite.

Contribution

It introduces the first unified agent framework for personalized fashion tasks that leverages multi-level negative feedback for adaptive refinement.

Findings

Outperforms baseline models in recommendation quality

Achieves higher style consistency and visual quality

Establishes a new benchmark for intelligent fashion systems

Abstract

The advancement of intelligent agents has revolutionized problem-solving across diverse domains, yet solutions for personalized fashion styling remain underexplored, which holds immense promise for promoting shopping experiences. In this work, we present StyleTailor, the first collaborative agent framework that seamlessly unifies personalized apparel design, shopping recommendation, virtual try-on, and systematic evaluation into a cohesive workflow. To this end, StyleTailor pioneers an iterative visual refinement paradigm driven by multi-level negative feedback, enabling adaptive and precise user alignment. Specifically, our framework features two core agents, i.e., Designer for personalized garment selection and Consultant for virtual try-on, whose outputs are progressively refined via hierarchical vision-language model feedback spanning individual items, complete outfits, and try-on efficacy. Counterexamples are aggregated into negative prompts, forming a closed-loop mechanism that enhances recommendation quality. To assess the performance, we introduce a comprehensive evaluation suite encompassing style consistency, visual quality, face similarity, and artistic appraisal. Extensive experiments demonstrate StyleTailor's superior performance in delivering personalized designs and recommendations, outperforming strong baselines without negative feedback and establishing a new benchmark for intelligent fashion systems.