# Autonomous Fashion Outfit Composition via Unified Aesthetic Foresight Model

arXiv: https://arxiv.org/abs/2608.13888 · HTML: https://mlsanity.com/paper/2608.13888/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 30 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [Learning Fashion Compatibility with Bidirectional LSTMs](https://mlsanity.com/paper/1707.05691/) (2017)
- [Collaborative Large Language Model for Recommender Systems](https://mlsanity.com/paper/2311.01343/) (2023)
- [Learning Type-Aware Embeddings for Fashion Compatibility](https://mlsanity.com/paper/1803.09196/) (2018)
- [Fashion Outfit Complementary Item Retrieval](https://mlsanity.com/paper/1912.08967/) (2019)
- [Contrastive language and vision learning of general fashion concepts](https://mlsanity.com/paper/2204.03972/) (2022)
- [Learning Similarity Conditions Without Explicit Supervision](https://mlsanity.com/paper/1908.08589/) (2019)
- [Diffusion Models for Generative Outfit Recommendation](https://mlsanity.com/paper/2402.17279/) (2024)
- [Hierarchical Fashion Graph Network for Personalized Outfit Recommendation](https://mlsanity.com/paper/2005.12566/) (2020)
- [Dressing as a Whole: Outfit Compatibility Learning Based on Node-wise Graph Neural Networks](https://mlsanity.com/paper/1902.08009/) (2019)
- [OutfitTransformer: Learning Outfit Representations for Fashion Recommendation](https://mlsanity.com/paper/2204.04812/) (2022)
- [Scheduling Techniques for Liver Segmentation: ReduceLRonPlateau Vs OneCycleLR](https://mlsanity.com/paper/2202.06373/) (2022)
- [Learning to Synthesize Compatible Fashion Items Using Semantic Alignment and Collocation Classification: An Outfit Generation Framework](https://mlsanity.com/paper/2502.06827/) (2025)
- [Decoding Style: Efficient Fine-Tuning of LLMs for Image-Guided Outfit Recommendation with Preference](https://mlsanity.com/paper/2409.12150/) (2024)
- [FashionDPO:Fine-tune Fashion Outfit Generation Model using Direct Preference Optimization](https://mlsanity.com/paper/2504.12900/) (2025)
- [Hybrid-Hierarchical Fashion Graph Attention Network for Compatibility-Oriented and Personalized Outfit Recommendation](https://mlsanity.com/paper/2508.11105/) (2025)
- [Fashion Image-to-Image Translation for Complementary Item Retrieval](https://mlsanity.com/paper/2408.09847/) (2024)

## Abstract (arXiv)

Fashion Outfit Composition (FOC) requires sequentially assembling fashion items into a stylistically cohesive ensemble. Existing works struggle to model this step-by-step process effectively, primarily because they fail to jointly optimize the two critical capabilities required for FOC: intermediate outfit value evaluation and complementary item prediction. This structural disconnect, compounded by the severe sparsity of step-wise aesthetic signals, leaves them without a mechanism to autonomously determine when to stop the composition process. To address these challenges, we introduce the Unified Aesthetic Foresight Model (UAFM), which seamlessly unifies both capabilities into a dual-head architecture over a shared backbone. Crucially, we formulate FOC as a deterministic Markov Decision Process and optimize a value head via a post-decision state temporal difference (TD) objective to recursively backpropagate sparse terminal aesthetic rewards to intermediate states. This enables the model to accurately estimate the potential of a partial outfit evolving into a compatible ensemble. Consequently, UAFM can compute step-wise marginal aesthetic gains for dynamic termination. Furthermore, this unified design strictly aligns aesthetic evaluation and complementary item prediction within a shared representation manifold, intrinsically regularizing the combinatorial search space. Extensive experiments on the Polyvore-Outfits dataset demonstrate that UAFM establishes a new state-of-the-art across both FOC and conventional fashion tasks. Extensive ablation studies further confirm that our post-decision state TD formulation provides the aesthetic value prediction necessary for autonomous dynamic termination, while validating the synergistic benefits of our unified dual-head architecture.