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

Excerpt

arXiv:2606.26899v2 Announce Type: replace Abstract: 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 con