# VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion

arXiv: https://arxiv.org/abs/2609.12793 · HTML: https://mlsanity.com/paper/2609.12793/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 40 references that are in the map.
- First posted on arXiv 11 Sep 2026.

## Builds on (most cited first, 3 of 3)

- [There and Back Again: Revisiting Backpropagation Saliency Methods](https://mlsanity.com/paper/2004.02866/) (2020)
- [Applications of deep learning in stock market prediction: recent progress](https://mlsanity.com/paper/2003.01859/) (2020)
- [A Comparative Analysis of Forecasting Financial Time Series Using ARIMA, LSTM, and BiLSTM](https://mlsanity.com/paper/1911.09512/) (2019)

## Built on by (most cited first, 1 of 1)

- [DSTNet: Dynamic Spectral Trajectory Network for Causal Multi-Horizon Financial Forecasting](https://mlsanity.com/paper/2610.09654/) (2026)

## Abstract (arXiv)

Stock price prediction remains challenging due to the non-stationary and noisy nature of financial time series. Existing deep learning models often rely on rigid decision-level fusion, ad hoc hyperparameter tuning, and compressed final-layer outputs, causing information loss, overfitting, and limited cross-market generalization. We propose VertiFuseX, a hybrid LSTM architecture using penultimate-layer vertical fusion of multi-scale temporal representations. VertiFuseX stacks and reweights penultimate features from LSTM, Bi-LSTM, and St-LSTM branches, integrates a parallel DNN stream, and jointly optimizes all components via backpropagation under a fixed hyperparameter configuration. This preserves richer intermediate temporal information across scales. Evaluated on 15 years (2010-2024) of closing prices from 10 global equity indices using strict chronological out-of-sample testing with the final 365 trading days held out, VertiFuseX achieves 30-54% MAPE reductions and over 40% improvements in MAE and RMSE versus LSTM-based baselines, and outperforms seven state-of-the-art models across 33 metric-dataset comparisons. Ablation studies confirm penultimate-layer fusion drives these gains over final-layer fusion and decision-level ensembling. Gradient-based saliency analysis shows consistent emphasis on mid-range dependencies at lags 9-15 days. Economic validation via algorithmic trading simulation under extreme market regimes shows reduced maximum drawdowns and superior risk-adjusted returns. With 675k parameters, a 2.6 MB memory footprint, and 1.5 ms/sample inference latency, VertiFuseX offers a lightweight, interpretable, deployment-ready framework for robust financial forecasting.