🤖 AI Summary
This study addresses the inability of traditional single-asset forecasting models to capture cross-firm event propagation, which results in delayed market responses. To overcome this limitation, the authors propose a heterogeneous Rust-Python streaming architecture that maps news in real time onto continuous-time heterogeneous graphs and models inter-firm influence through a dynamic attention mechanism. Key innovations include zero-copy streaming parsing, node-level continuous-time LSTMs, a multivariate Neural Hawkes process, bilinear latent projection, and adaptive neighborhood pruning. The system achieves an end-to-end latency of only 13 ms. Evaluated on the FNSPID corpus, it improves next-day return prediction accuracy at the 90th percentile by 1.70× over random guessing and by 3.36× over industry baselines. Ablation studies confirm that graph topology is the sole source of cross-firm predictive signals.
📝 Abstract
Per-ticker forecasting models dominate financial time-series work yet remain blind to cross-company propagation: a foundry disruption in Taiwan does not register in a single-asset model until Apple's own price has already moved. To address this limitation, we introduce a heterogeneous Rust-Python streaming architecture that maps cross-company attention as a continuous-time graph driven directly from text. We show that on the ingestion side, a zero-copy Rust edge parses news records in $\sim$100 ns and scans the target equity universe in $\sim$1.2 $μ$s. On the inference end, a multivariate Neural Hawkes Process featuring per-node continuous-time LSTM states and a bilinear latent projection propagates directed excitation, while an adaptive pruning rule bounds the computational cost of dynamic neighborhood updates. Combining these stages, we demonstrate an end-to-end processing latency of $\sim$13 ms per incoming news record on a single commodity CPU. Evaluated on a one-month temporal holdout of the FNSPID corpus (638 articles across 47 tickers), the system delivers a $1.70\times$ precision lift over random at the 90th-percentile next-day return threshold, and $3.36\times$ over a same-sector baseline. Crucially, removing the graph topology collapses precision to zero, confirming that the dynamic attention network is the sole driver of cross-company signal in this architecture.