Search papers, labs, and topics across Lattice.
This paper introduces EvtGraph, a novel framework that optimizes the representation of multimodal temporal data by aligning computation with temporal salience through event-adaptive compression and node budget constraints. By transforming dense sequences into structured computations over salient events, EvtGraph effectively reduces complexity while maintaining critical transitions, demonstrating a strong performance-efficiency trade-off. Experimental results on multimodal clinical and cross-domain benchmarks reveal that EvtGraph consistently outperforms traditional Transformer-based and recurrent models, highlighting its practical advantages in high-redundancy temporal data scenarios.
Event-adaptive compression allows EvtGraph to outperform traditional models while maintaining efficiency, proving that less can be more in temporal data representation.
Multimodal temporal data are inherently irregular and uneven in information density, yet most models rely on uniform discretization, leading to inefficient representations. We propose \textbf{EvtGraph}, a unified framework that aligns computation with temporal salience under explicit budget constraints. EvtGraph reparameterizes sequences into event-level tokens via event-adaptive compression (EAMC), selects a compact subset with a node budget (NBC), and performs temporally constrained sparse graph reasoning (T2SG). This transforms dense sequences into structured computation over salient events, reducing complexity while preserving critical transitions. We show that this design provides a practical mechanism for allocating representational capacity under a fixed budget, yielding a consistent performance--efficiency trade-off, where a small budget is often sufficient in practice. Experiments on multimodal clinical (MIMIC-IV + CXR) and cross-domain benchmarks demonstrate that EvtGraph outperforms both Transformer-based and recurrent baselines while significantly improving efficiency. These results suggest that budget-constrained event-centric representation provides a general paradigm for learning from high-redundancy temporal data.
Tsinghua AI