Tag
Hallo4D is a model-agnostic framework that leverages large multimodal language models to detect and correct spatial and temporal hallucinations in 3D and 4D generation, improving consistency across viewpoints and time without requiring retraining.
This paper introduces Evidence-Backed Video Question Answering (E-VQA), a new task requiring models to output both semantic answers and precise spatio-temporal evidence like tracked object segmentation masklets. The authors create a human-verified benchmark and a scalable training dataset, showing significant improvements over baselines.
STAGformer introduces a spatio-temporal agent graph transformer with linear complexity for bike-sharing demand forecasting, outperforming baselines on NYC and Chicago datasets.
Introduces AnyGroundBench, a domain-adaptation benchmark for spatio-temporal video grounding, evaluating 15 VLMs across five specialized domains and finding current models fail in zero-shot and in-context learning adaptation.
This paper presents a deep learning approach using a spatio-temporal graph neural network (MTGNN) to reconstruct GRACE terrestrial water storage anomalies back to 1940 for South America, achieving high accuracy and outperforming previous methods with fewer predictors.
This paper proposes MVG-KAN, a multi-view model integrating periodic-residual decomposition, a Geo-Wind Graph for wind-aware spatial dependencies, and a temporal KAN head for PM2.5 forecasting, achieving MAE 14.09 on Beijing data.
Selective Synergistic Learning (SSync) improves video object-centric learning by selectively distilling reliable cues via pseudo-labeling and transitive merging, avoiding error propagation from indiscriminate dense alignment.
Proposes a node-level spectral energy formulation for detecting camouflaged anomalies in graphs, extending to spatio-temporal settings with energy-driven message passing. Demonstrates effectiveness on large-scale benchmarks.
This paper proposes using the Ensemble Score Filter (EnSF), a score-based diffusion data assimilation method, to correct forecasts from a pretrained spatio-temporal energy consumption model using noisy partial observations. Numerical experiments show EnSF significantly improves state estimation over open-loop propagation and outperforms the Ensemble Kalman Filter under nonlinear observations.
Proposes a Global-Local Graph Attention Network (GLGAT) with pairwise encoding and event-based adjacency matrix for traffic forecasting, effectively capturing spatio-temporal correlations and achieving competitive performance on real-world datasets.