Drift-Aware Multimodal User Representation Learning via Multi-Scale Temporal Modeling and Sparse Mixture-of-Experts

• URL: http://arxiv.org/abs/2608.25773v1
• Date: Wed, 26 Aug 2026 13:15:17 GMT
• Title: Drift-Aware Multimodal User Representation Learning via Multi-Scale Temporal Modeling and Sparse Mixture-of-Experts
• Abstract summary: We propose DUMoE, a unified framework for drift-aware multimodal user representation learning.<n>Our model consists of (i) a temporal dynamics-aware backbone that captures and integrates static profiles, short-term behavioral signals, and long-term dependencies into a coherent representation, and (ii) a sparse mixture-of-experts (MoE) interest adapter that disentangles multiple latent interests via expert specialization and adaptive routing.<n>Experiments on real-world social media datasets show that DUMoE consistently outperforms state-of-the-art methods on both user interest prediction and interaction prediction tasks.
• Score: 8.624467983444255
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Understanding user preferences from noisy and temporally evolving social media behaviors is fundamentally challenging due to interest drift, where user preferences shift across time and exhibit both multi-scale temporal patterns and diverse co-existing interests. To address this, we propose DUMoE, a unified framework for drift-aware multimodal user representation learning. Our model consists of (i) a temporal dynamics-aware backbone that captures and integrates static profiles, short-term behavioral signals, and long-term dependencies into a coherent representation, and (ii) a sparse mixture-of-experts (MoE) interest adapter that disentangles multiple latent interests via expert specialization and adaptive routing. Each expert models a distinct interest subspace, while a gating network dynamically selects and aggregates a sparse subset of relevant experts for each user. To enable stable and effective optimization, we further introduce a three-stage training strategy that decouples backbone learning, expert specialization, and gating optimization. Extensive experiments on real-world social media datasets show that DUMoE consistently outperforms state-of-the-art methods on both user interest prediction and interaction prediction tasks.

Related papers

• SLSREC: Self-Supervised Contrastive Learning for Adaptive Fusion of Long- and Short-Term User Interests [22.535188624090353]
  SLSRec is a Session-based model with the fusion of Long- and Short-term Recommendations.<n>A contrastive learning strategy is introduced to ensure accurate calibration of long- and short-term interest representations.<n>Experiments on three public benchmark datasets demonstrate that SLSRec consistently outperforms state-of-the-art models.
  arXiv  Detail & Related papers  (2026-04-06T08:49:46Z)
• Modeling Stage-wise Evolution of User Interests for News Recommendation [11.474027863114854]
  We propose a unified framework that learns user preferences from both global and local temporal perspectives.<n>A global preference modeling component captures long-term collaborative signals from the overall interaction graph.<n>A local preference modeling component partitions historical interactions into stage-wise temporal subgraphs to represent short-term dynamics.
  arXiv  Detail & Related papers  (2026-03-11T06:42:52Z)
• CAMMSR: Category-Guided Attentive Mixture of Experts for Multimodal Sequential Recommendation [23.478610632707728]
  We propose a Category-guided Attentive Mixture of Experts model for Multimodal Sequential Recommendation.<n>At its core, CAMMSR introduces a category-guided attentive mixture of experts module, which learns specialized item representations from multiple perspectives.<n>Experiments on four public datasets demonstrate that CAMMSR consistently outperforms state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-03-04T17:39:35Z)
• Mixture of Sequence: Theme-Aware Mixture-of-Experts for Long-Sequence Recommendation [48.98369214412951]
  We propose a model-agnostic MoE approach that achieves accurate predictions by extracting theme-specific and multi-scale subsequences from noisy raw user sequences.<n> Experimental results demonstrate that MoS consistently achieves the SOTA performance while introducing fewer FLOPs.
  arXiv  Detail & Related papers  (2026-03-01T23:20:48Z)
• Spatiotemporal-Untrammelled Mixture of Experts for Multi-Person Motion Prediction [53.555201955973104]
  Comprehensively flexibly capturing the complex-temporal dependencies of human motion is critical for multi-person motion.<n>Existing methods grapple with two primary limitations.<n>High computational costs stemming from time of conventional attention.<n>Our model incorporates four distinct types oftemporal experts, each specializing in capturing different spatial or temporal dependencies.
  arXiv  Detail & Related papers  (2025-12-25T15:01:19Z)
• FusAD: Time-Frequency Fusion with Adaptive Denoising for General Time Series Analysis [92.23551599659186]
  Time series analysis plays a vital role in fields such as finance, healthcare, industry, and meteorology.<n>FusAD is a unified analysis framework designed for diverse time series tasks.
  arXiv  Detail & Related papers  (2025-12-16T04:34:27Z)
• Hierarchical Time-Aware Mixture of Experts for Multi-Modal Sequential Recommendation [19.47124940518026]
  We propose a Hierarchical time-aware Mixture of experts for multi-modal Sequential Recommendation (HM4SR)<n>First MoE, named Interactive MoE, extracts essential user interest-related information from the multi-modal data of each item.<n>Second MoE, termed Temporal MoE, captures user dynamic interests by introducing explicit temporal embeddings from timestamps in modality encoding.
  arXiv  Detail & Related papers  (2025-01-24T06:26:50Z)
• LLM-based Bi-level Multi-interest Learning Framework for Sequential Recommendation [54.396000434574454]
  We propose a novel multi-interest SR framework combining implicit behavioral and explicit semantic perspectives.<n>It includes two modules: the Implicit Behavioral Interest Module and the Explicit Semantic Interest Module.<n>Experiments on four real-world datasets validate the framework's effectiveness and practicality.
  arXiv  Detail & Related papers  (2024-11-14T13:00:23Z)
• DeepInteraction++: Multi-Modality Interaction for Autonomous Driving [80.8837864849534]
  We introduce a novel modality interaction strategy that allows individual per-modality representations to be learned and maintained throughout.<n>DeepInteraction++ is a multi-modal interaction framework characterized by a multi-modal representational interaction encoder and a multi-modal predictive interaction decoder.<n>Experiments demonstrate the superior performance of the proposed framework on both 3D object detection and end-to-end autonomous driving tasks.
  arXiv  Detail & Related papers  (2024-08-09T14:04:21Z)
• Learning Dual Dynamic Representations on Time-Sliced User-Item Interaction Graphs for Sequential Recommendation [62.30552176649873]
  We devise a novel Dynamic Representation Learning model for Sequential Recommendation (DRL-SRe) To better model the user-item interactions for characterizing the dynamics from both sides, the proposed model builds a global user-item interaction graph for each time slice. To enable the model to capture fine-grained temporal information, we propose an auxiliary temporal prediction task over consecutive time slices.
  arXiv  Detail & Related papers  (2021-09-24T07:44:27Z)
• MRIF: Multi-resolution Interest Fusion for Recommendation [0.0]
  This paper presents a multi-resolution interest fusion model (MRIF) that takes both properties of users' interests into consideration. The proposed model is capable to capture the dynamic changes in users' interests at different temporal-ranges, and provides an effective way to combine a group of multi-resolution user interests to make predictions.
  arXiv  Detail & Related papers  (2020-07-08T02:32:15Z)
• Sequential Recommender via Time-aware Attentive Memory Network [67.26862011527986]
  We propose a temporal gating methodology to improve attention mechanism and recurrent units. We also propose a Multi-hop Time-aware Attentive Memory network to integrate long-term and short-term preferences. Our approach is scalable for candidate retrieval tasks and can be viewed as a non-linear generalization of latent factorization for dot-product based Top-K recommendation.
  arXiv  Detail & Related papers  (2020-05-18T11:29:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.