Tag
This paper proposes a geometry-aware multi-support heterogeneous graph neural network for fine-scale rainfall field reconstruction, which fuses observations from point gauges, path-integrated microwave links, and gridded radar/satellite data. The method reduces RMSE by 23.2% over classical interpolation on Singapore data and shows greatest gains when the field is undersampled relative to its spatial correlation length.
This paper proposes HIA-GAT, a dual-stream heterogeneous graph attention network that integrates longitudinal and lateral vehicle interactions with a conflict-type-aware gating mechanism for frame-level traffic conflict risk prediction on freeways. Experiments on NGSIM datasets show improved risk-ranking performance, particularly for lateral conflicts, and provide interpretable per-vehicle conflict attribution.
Proposes a hierarchical semantic-constrained heterogeneous graph model for open-vocabulary audio-visual event localization, addressing cross-modal consistency at multiple temporal scales and hierarchical semantic constraints between segment and video levels. Achieves state-of-the-art results on OV-AVEL benchmark.