MealRec: Multi-granularity Sequential Modeling via Hierarchical Diffusion Models for Micro-Video Recommendation

TL;DR

A Multi-granularity sequential modeling method via hierarchical diffusion models for micro-video Recommendation (MealRec), which simultaneously considers temporal correlations during preference modeling from intra- and inter-video perspectives, and designs the Noise-unconditional Preference Denoising (NPD) to recovers informative user preferences from corrupted states under the blind denoising.

Abstract

Micro-video recommendation aims to capture user preferences from the collaborative and context information of the interacted micro-videos, thereby predicting the appropriate videos. This target is often hindered by the inherent noise within multimodal content and unreliable implicit feedback, which weakens the correspondence between behaviors and underlying interests. While conventional works have predominantly approached such scenario through behavior-augmented modeling and content-centric multimodal analysis, these paradigms can inadvertently give rise to two non-trivial challenges: preference-irrelative video representation extraction and inherent modality conflicts. To address these issues, we propose a Multi-granularity sequential modeling method via hierarchical diffusion models for micro-video Recommendation (MealRec), which simultaneously considers temporal correlations during preference modeling from intra- and inter-video perspectives. Specifically, we first propose Temporal-guided Content Diffusion (TCD) to refine video representations under intra-video temporal guidance and personalized collaborative signals to emphasize salient content while suppressing redundancy. To achieve the semantically coherent preference modeling, we further design the Noise-unconditional Preference Denoising (NPD) to recovers informative user preferences from corrupted states under the blind denoising. Extensive experiments and analyses on four micro-video datasets from two platforms demonstrate the effectiveness, universality, and robustness of our MealRec, further uncovering the effective mechanism of our proposed TCD and NPD. The source code and corresponding dataset will be available upon acceptance.

Figures (7)

• Figure 1: Illustration of the proposed MealRec as a chef preparing a meal: it first leverages the "kitchenware" (DM) to sequentially slice and recombine "ingredients" (micro-videos) into "preferences-aware compositions" (video representations), and then applies another "tool" (DM) to professionally refine these "compositions", ultimately serving the delicious "meal" (recommendation results).
• Figure 2: Overall structure of our proposed MealRec.
• Figure 3: Ablation analysis of MealRec across four datasets. Generally, each component within our MealRec is effective.
• Figure 4: Parameter sensitivity analysis about (a) & (b): The comparison between the loss weight $\lambda_{\mathrm{T}}$ and diffusion step $T^{\mathrm{t}}$ of TCD; (c) & (d): The comparison between the loss weight $\lambda_{\mathrm{T}}$ and diffusion step $T^{\mathrm{t}}$ of NPD. Here, darker colors denote higher scores.
• Figure 5: Performance of MealRec under different levels of Gaussian noise ($\sigma$) added to the pre-extracted visual features.