# Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization

arXiv: https://arxiv.org/abs/2606.26899 · HTML: https://mlsanity.com/paper/2606.26899/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Diffusion recommenders, one of 14 directions in Diffusion models.
- Builds on 35 of its 38 references that are in the map.
- First posted on arXiv 25 Jun 2026.

## Placed in

- Diffusion models / [Diffusion recommenders](https://mlsanity.com/diffusion/diffusion-recommenders/)
- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 30 of 35)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Training Diffusion Models with Reinforcement Learning](https://mlsanity.com/paper/2305.13301/) (2023)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [Autoregressive Image Generation without Vector Quantization](https://mlsanity.com/paper/2406.11838/) (2024)
- [Precise Zero-Shot Dense Retrieval without Relevance Labels](https://mlsanity.com/paper/2212.10496/) (2022)
- [DanceGRPO: Unleashing GRPO on Visual Generation](https://mlsanity.com/paper/2505.07818/) (2025)
- [Autoregressive Entity Retrieval](https://mlsanity.com/paper/2010.00904/) (2020)
- [Flow Matching on General Geometries](https://mlsanity.com/paper/2302.03660/) (2023)
- [Transformer Memory as a Differentiable Search Index](https://mlsanity.com/paper/2202.06991/) (2022)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization](https://mlsanity.com/paper/2310.03708/) (2023)
- [Diffusion Recommender Model](https://mlsanity.com/paper/2304.04971/) (2023)
- [CompoDiff: Versatile Composed Image Retrieval With Latent Diffusion](https://mlsanity.com/paper/2303.11916/) (2023)
- [Generate What You Prefer: Reshaping Sequential Recommendation via Guided Diffusion](https://mlsanity.com/paper/2310.20453/) (2023)
- [Improving Query Representations for Dense Retrieval with Pseudo Relevance Feedback](https://mlsanity.com/paper/2108.13454/) (2021)

## Abstract (arXiv)

Embedding-based retrieval typically returns highest-scoring items, but many production scenarios require items that satisfy a target attribute while preserving a fine-grained pattern expressed by seed examples. We formalize this as pattern-preserving attribute retrieval. Standard approaches fail: averaging seeds preserves the pattern but misses the attribute; global attribute retrieval drifts to unrelated patterns. We approach the task with continuous generative retrieval, where a model reads item-embedding sequences and generates query embeddings for nearest-neighbor search. We propose MORPH: Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization, a staged framework with large-scale raw-sequence pretraining, Metric-Ordered Sequence (MOS) training, and final HPPO alignment. MOS construction turns sparse online metric labels into in-pattern trajectories; MOS CPT/SFT then trains the generator through shared multi-domain continuation pretraining and domain-specific tail-centroid supervised fine-tuning. HPPO uses a hybrid pool of static and policy-generated candidate embeddings, labels them with true online intersection metrics, applies iterated preference optimization, and employs a Pareto pair filter to exclude winners that lower pattern purity. Across four large-scale attribute domains under strict item- and pattern-holdout protocols, MOS training improves the primary intersection metric over a strong pretrained generative retriever in every domain-split cell, and the complete Pareto-filtered HPPO procedure improves it further, with paired-bootstrap-significant gains on seven of the eight cells - the exception being the D4 pattern-holdout split. Ablations confirm that the Pareto pair filter improves the attribute-pattern tradeoff on D1-D3, and that hybrid static/policy candidates are complementary.