mlsanityData to 7 Oct 2026

Efficient, Property-Aligned Fan-Out Retrieval via RL-Compiled Diffusion

arXiv 2603.06397 ↗First posted 6 Mar 2026cs.IR cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Many modern retrieval problems are set-valued: given a broad intent, the system must return a collection of results that optimizes higher-order properties (e.g., diversity, coverage, complementarity, coherence) while remaining grounded with respect to a fixed database. Set-valued objectives are typically non-decomposable and are not captured by existing supervised (query, content) datasets which only prioritize top-1 retrieval. Consequently, fan-out retrieval is often employed to generate diverse subqueries to retrieve item sets. While reinforcement learning (RL) can optimize set-level objectives via interaction, deploying an RL-tuned LLM for fan-out retrieval is prohibitively expensive at inference time. Conversely, diffusion-based generative retrieval enables efficient single-pass fan-out in embedding space, but requires objective-aligned training targets. To address these issues, we propose R4T (Retrieve-for-Train), which uses RL once as an objective transducer in a three-step process: (i) train a fan-out LLM with composite set-level rewards, (ii) synthesize objective-consistent training pairs, and (iii) train a lightweight diffusion retriever to model the conditional distribution of set-valued outputs. Across large-scale fashion and music benchmarks consisting of curated item sets, we show that R4T improves retrieval quality relative to strong baselines while reducing query-time fan-out latency by an order of magnitude.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 39 of its 52 references that are in the map. First posted on arXiv 6 Mar 2026.

Builds on

39 of 52 references.

References · 39

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  4. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  5. 2025Qwen3 Technical Reportcited by 8,195
  6. 2022Scalable Diffusion Models with Transformerscited by 5,931
  7. 2022Classifier-Free Diffusion Guidancecited by 5,717
  8. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  9. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  10. 2025Gemma 3 Technical Reportcited by 1,686
  11. 2022Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentationcited by 707
  12. 2025The Entropy Mechanism of Reinforcement Learning for Reasoning Language Modelscited by 463
27 more
  1. 2022The Vendi Score: A Diversity Evaluation Metric for Machine Learningcited by 299
  2. 2022Matryoshka Representation Learningcited by 268
  3. 2022Transformer Memory as a Differentiable Search Indexcited by 242
  4. 2019Document Expansion by Query Predictioncited by 225
  5. 2023Query Rewriting for Retrieval-Augmented Large Language Modelscited by 211
  6. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 175
  7. 2023Machine Learning for Synthetic Data Generation: A Reviewcited by 118
  8. 2022Autoregressive Search Engines: Generating Substrings as Document Identifierscited by 112
  9. 2022A Neural Corpus Indexer for Document Retrievalcited by 107
  10. 2022MuLan: A Joint Embedding of Music Audio and Natural Languagecited by 85
  11. 2017Learning Fashion Compatibility with Bidirectional LSTMscited by 69
  12. 2025DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learningcited by 62
  13. 2023How Does Generative Retrieval Scale to Millions of Passages?cited by 47
  14. 2023Learning to Rank in Generative Retrievalcited by 46
  15. 2017Query expansion techniques for information retrieval: A surveycited by 44
  16. 2025s3: You Don't Need That Much Data to Train a Search Agent via RLcited by 41
  17. 2023Cross-Modal Retrieval: A Systematic Review of Methods and Future Directionscited by 40
  18. 2023Scalable and Effective Generative Information Retrievalcited by 38
  19. 2023When do you need Chain-of-Thought Prompting for ChatGPT?cited by 34
  20. 2025On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoningcited by 30
  21. 2025Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skillscited by 19
  22. 2023Multiview Identifiers Enhanced Generative Retrievalcited by 16
  23. 2023Generative Slate Recommendation with Reinforcement Learningcited by 12
  24. 2024Generative Retrieval as Multi-Vector Dense Retrievalcited by 8
  25. 2025TROLL: Trust Regions improve Reinforcement Learning for Large Language Modelscited by 7
  26. 2024Prompt-to-Slate: Diffusion Models for Prompt-Conditioned Slate Generationcited by 6
  27. 2025GD-Retriever: Controllable Generative Text-Music Retrieval with Diffusion Modelscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Optimizing Retrieval for RAG via Reinforcement Learning
  2. 2025Annotation-Free Reinforcement Learning Query Rewriting via Verifiable Search Reward
  3. 2026Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization
  4. 2026It Takes Two to Match: Co-Evolving Generative Retriever with Reinforcement Learning
  5. 2026Reinforcement Fine-Tuning for History-Aware Dense Retriever in RAG
  6. 2025Domain-Aware RAG: MoL-Enhanced RL for Efficient Training and Scalable Retrieval
  7. 2026DARE: Aligning LLM Agents with the R Statistical Ecosystem via Distribution-Aware Retrieval
  8. 2026DSL-R1: From SQL to DSL for Training Retrieval Agents across Structured and Unstructured Data with Reinforcement Learning