# When Does Forecasting Reveal Temporal Structure? A Stability Analysis of Time-Series Structural Selection

arXiv: https://arxiv.org/abs/2608.10433 · HTML: https://mlsanity.com/paper/2608.10433/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 26 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Builds on (most cited first, 13 of 13)

- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [Underspecification Presents Challenges for Credibility in Modern Machine Learning](https://mlsanity.com/paper/2011.03395/) (2020)
- [Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations](https://mlsanity.com/paper/1703.03717/) (2017)
- [Detecting causal associations in large nonlinear time series datasets](https://mlsanity.com/paper/1702.07007/) (2017)
- [Predictive Multiplicity in Classification](https://mlsanity.com/paper/1909.06677/) (2019)
- [Discovering contemporaneous and lagged causal relations in autocorrelated nonlinear time series datasets](https://mlsanity.com/paper/2003.03685/) (2020)
- [TimeX++: Learning Time-Series Explanations with Information Bottleneck](https://mlsanity.com/paper/2405.09308/) (2024)
- [Rethinking Channel Dependence for Multivariate Time Series Forecasting: Learning from Leading Indicators](https://mlsanity.com/paper/2401.17548/) (2024)
- [DSPR: Dual-Stream Physics-Residual Networks for Trustworthy Industrial Time Series Forecasting](https://mlsanity.com/paper/2604.07393/) (2026)
- [The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations](https://mlsanity.com/paper/2607.16236/) (2026)
- [Information Bottleneck Learning for Faithful Time Series Forecasting Explanations](https://mlsanity.com/paper/2607.28124/) (2026)

## Abstract (arXiv)

Forecast accuracy is often used as a proxy for temporal structure discovery, but predictive performance and structural identifiability are not equivalent. Different temporal mechanisms can achieve similar forecast errors, while small forecast differences may still contain sufficient information for recovery. In this work, we study when forecast-only structural selection can be trusted. We show that a vanishing forecast margin does not necessarily imply structural ambiguity, and establish a stability perspective that evaluates structural separation relative to uncertainty in the selection objective. This perspective provides both a sufficient condition for reliable selection and a continuous measure of selection difficulty. Experiments across controlled and end-to-end settings demonstrate that forecast margin alone is insufficient, while the proposed stability measure better characterizes when forecast-based structural selection succeeds or fails. Our results suggest that predictive accuracy should be treated as evidence for structure discovery only when its separation is sufficiently robust.