aiqbench analysis
The paper argues that fixed, globally shared fusion and joint loss weighting underperform on urban data, where the value of each signal (mobility, points of interest, land use) varies by region and objectives conflict during training. PRISM addresses this with sparse mixture-of-experts routing for region-specific view weighting and a PID-inspired controller to damp gradient conflicts. The reported gains are described as "competitive" with strong baselines, so readers should check the actual margins and the stability claims against the full experiments before treating the method as a durable improvement.
Rethinking Multi-View Fusion for Urban Representation Learning
arXiv:2610.31850v1 Announce Type: new Abstract: Urban region representation learning encodes heterogeneous urban signals, such as mobility flows, points of interest, and land-use patterns, into unified embeddings for downstream tasks, including check-in prediction, crime forecasting, and service demand estimation. Existing approaches rely on globally shared fusion and joint optimization, implicitly assuming uniform modality importance and naturally balanced multi-objective training. However, urban data is inherently heterogeneous. Our empirical analysis reveals two structural issues: (i) modality utility varies markedly across regions, causing static fusion to overemphasize signals from certain modalities; and (ii) heterogeneous objectives exhibit persistent gradient imbalance, causing representational bias and unstable training. We present PRISM, a self-adjusting framework for adaptive multi-view learning. PRISM adapts sparse MoE routing to learn region-specific view allocations, while a PID (Proportional Integral Derivative)-inspired regulation mechanism mitigates gradient conflicts during optimization. Experiments on three large-scale urban datasets show that PRISM achieves competitive performance compared with strong multi-view baselines, while providing more stable optimization behavior and more balanced region representations.
Abstract:Urban region representation learning encodes heterogeneous urban signals, such as mobility flows, points of interest, and land-use patterns, into unified embeddings for downstream tasks, including check-in prediction, crime forecasting, and service demand estimation. Existing approaches rely on globally shared fusion and joint optimization, implicitly assuming uniform modality importance and naturally balanced multi-objective training. However, urban data is inherently heterogeneous. Our empirical analysis reveals two structural issues: (i) modality utility varies markedly across regions, causing static fusion to overemphasize signals from certain modalities; and (ii) heterogeneous objectives exhibit persistent gradient imbalance, causing representational bias and unstable training. We present PRISM, a self-adjusting framework for adaptive multi-view learning. PRISM adapts sparse MoE routing to learn region-specific view allocations, while a PID (Proportional Integral Derivative)-inspired regulation mechanism mitigates gradient conflicts during optimization. Experiments on three large-scale urban datasets show that PRISM achieves competitive performance compared with strong multi-view baselines, while providing more stable optimization behavior and more balanced region representations.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.31850 [cs.LG]
(or arXiv:2610.31850v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.31850 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 254 words
Related coverage
Rethinking Implicit Spatial Representation in Visuomotor Policy Learning
arXiv · Jun 16, 2026
Rethinking Molecular Text Representations for LLMs: An Empirical Study
arXiv · Jun 3, 2026
From Scalars to Time Series: Rethinking Implicit Neural Representations for Time-Varying Volumetric Data
arXiv · Jul 24, 2026
ScaleMPA: Rethinking Scalable RRT* Acceleration With a Grid-Native Representation
arXiv · Sep 22, 2026