# Preference Flow Matching with Spectral Factorization for Micro-video Recommendation

arXiv: https://arxiv.org/abs/2608.26579 · HTML: https://mlsanity.com/paper/2608.26579/

## 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 22 of its 33 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Placed in

- Diffusion models / [Diffusion recommenders](https://mlsanity.com/diffusion/diffusion-recommenders/)

## Builds on (most cited first, 22 of 22)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [DiffuRec: A Diffusion Model for Sequential Recommendation](https://mlsanity.com/paper/2304.00686/) (2023)
- [Frequency Enhanced Hybrid Attention Network for Sequential Recommendation](https://mlsanity.com/paper/2304.09184/) (2023)
- [A Content-Driven Micro-Video Recommendation Dataset at Scale](https://mlsanity.com/paper/2309.15379/) (2023)
- [Deconfounding Duration Bias in Watch-time Prediction for Video Recommendation](https://mlsanity.com/paper/2206.06003/) (2022)
- [Plug-in Diffusion Model for Sequential Recommendation](https://mlsanity.com/paper/2401.02913/) (2024)
- [Conditional Denoising Diffusion for Sequential Recommendation](https://mlsanity.com/paper/2304.11433/) (2023)
- [Spectrum-based Modality Representation Fusion Graph Convolutional Network for Multimodal Recommendation](https://mlsanity.com/paper/2412.14978/) (2024)
- [A Large-scale Dataset with Behavior, Attributes, and Content of Mobile Short-video Platform](https://mlsanity.com/paper/2502.05922/) (2025)
- [Flow Matching for Collaborative Filtering](https://mlsanity.com/paper/2502.07303/) (2025)
- [Efficient and Effective Adaptation of Multimodal Foundation Models in Sequential Recommendation](https://mlsanity.com/paper/2411.02992/) (2024)
- [Understanding and Modeling Passive-Negative Feedback for Short-video Sequential Recommendation](https://mlsanity.com/paper/2308.04086/) (2023)
- [FITMM: Adaptive Frequency-Aware Multimodal Recommendation via Information-Theoretic Representation Learning](https://mlsanity.com/paper/2601.22498/) (2026)
- [Dynamic Multimodal Fusion via Meta-Learning Towards Micro-Video Recommendation](https://mlsanity.com/paper/2501.07110/) (2025)
- [Large Language Model Empowered Recommendation Meets All-domain Continual Pre-Training](https://mlsanity.com/paper/2504.08949/) (2025)
- [FlowRec: Prior-Informed Flow Matching for Efficient Sequential Recommendation Generation](https://mlsanity.com/paper/2508.17618/) (2025)
- [Frequency Selective Augmentation for Video Representation Learning](https://mlsanity.com/paper/2204.03865/) (2022)
- [Short Video Segment-level User Dynamic Interests Modeling in Personalized Recommendation](https://mlsanity.com/paper/2504.04237/) (2025)
- [Frame2Freq: Spectral Adapters for Fine-Grained Video Understanding](https://mlsanity.com/paper/2602.18977/) (2026)
- [Exploiting Fine-Grained Skip Behaviors for Micro-Video Recommendation](https://mlsanity.com/paper/2504.03107/) (2025)

## Abstract (arXiv)

Micro-video recommendation aims to infer user preferences from historical interactions and multimodal video content, thereby identifying the next video of interest. However, prevailing methods compress frame sequences into a single holistic representation, entangling the stable visual semantics and the evolving dynamics that jointly shape user preferences. Meanwhile, diffusion- and flow matching-based recommenders condition their generation process solely on coarse behavioral context, leaving its internal temporal structure outside preference formation. We therefore propose PrismRec, a Preference Flow Matching framework with Spectral Factorization for Micro-video Recommendation. Analogous to a prism that disperses white light into its constituent spectrum, PrismRec devises Spectral Semantic Factorization (SSF) to derive complementary static semantic and dynamic factors from frame-level representations via a prior-guided learnable frequency mask in the temporal frequency domain. Then, it proposes Context-Calibrated Preference Matching (CPM) to weigh them with each user's specific sensitivity and inject the calibrated context as a structured condition to steer the matching trajectory toward the target representation, making video content as an intrinsic driver of preference formation rather than auxiliary side information. Experiments on four datasets from two platforms show that PrismRec surpasses the SOTA baseline by up to 22.65%, with the lowest inference cost and peak memory among the compared methods.