# On Identifying Why and When Foundation Models Perform Well on Time-Series Forecasting Using Automated Explanations and Rating

arXiv: https://arxiv.org/abs/2508.20437 · HTML: https://mlsanity.com/paper/2508.20437/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 46 references that are in the map.
- First posted on arXiv 28 Aug 2025.

## Builds on (most cited first, 18 of 18)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI](https://mlsanity.com/paper/1910.10045/) (2019)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [One Fits All:Power General Time Series Analysis by Pretrained LM](https://mlsanity.com/paper/2302.11939/) (2023)
- [Unified Training of Universal Time Series Forecasting Transformers](https://mlsanity.com/paper/2402.02592/) (2024)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [Temporal Pattern Attention for Multivariate Time Series Forecasting](https://mlsanity.com/paper/1809.04206/) (2018)
- [Towards a Rigorous Evaluation of XAI Methods on Time Series](https://mlsanity.com/paper/1909.07082/) (2019)
- [Do We Really Need Deep Learning Models for Time Series Forecasting?](https://mlsanity.com/paper/2101.02118/) (2021)
- [TS-MULE: Local Interpretable Model-Agnostic Explanations for Time Series Forecast Models](https://mlsanity.com/paper/2109.08438/) (2021)
- [EA-LSTM: Evolutionary Attention-based LSTM for Time Series Prediction](https://mlsanity.com/paper/1811.03760/) (2018)
- [Can Explainable AI Explain Unfairness? A Framework for Evaluating Explainable AI](https://mlsanity.com/paper/2106.07483/) (2021)
- [Retrieval-augmented Large Language Models for Financial Time Series Forecasting](https://mlsanity.com/paper/2502.05878/) (2025)
- [Rating Sentiment Analysis Systems for Bias through a Causal Lens](https://mlsanity.com/paper/2302.02038/) (2023)
- [The Effect of Human v/s Synthetic Test Data and Round-tripping on Assessment of Sentiment Analysis Systems for Bias](https://mlsanity.com/paper/2401.12985/) (2024)
- [Rating Multi-Modal Time-Series Forecasting Models (MM-TSFM) for Robustness Through a Causal Lens](https://mlsanity.com/paper/2406.12908/) (2024)
- [A Dataset and Baseline Approach for Identifying Usage States from Non-Intrusive Power Sensing With MiDAS IoT-based Sensors](https://mlsanity.com/paper/2209.00987/) (2022)

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

- [The Spectrum Is Not Enough: When Context Helps Time-Series Forecasting](https://mlsanity.com/paper/2607.13006/) (2026)

## Abstract (arXiv)

Time-series forecasting models (TSFM) have evolved from classical statistical methods to sophisticated foundation models, yet understanding why and when these models succeed or fail remains challenging. Despite this known limitation, time series forecasting models are increasingly used to generate information that informs real-world actions with equally real consequences. Understanding the complexity, performance variability, and opaque nature of these models then becomes a valuable endeavor to combat serious concerns about how users should interact with and rely on these models' outputs. This work addresses these concerns by combining traditional explainable AI (XAI) methods with Rating Driven Explanations (RDE) to assess TSFM performance and interpretability across diverse domains and use cases. We evaluate four distinct model architectures: ARIMA, Gradient Boosting, Chronos (time-series specific foundation model), Llama (general-purpose; both fine-tuned and base models) on four heterogeneous datasets spanning finance, energy, transportation, and automotive sales domains. In doing so, we demonstrate that feature-engineered models (e.g., Gradient Boosting) consistently outperform foundation models (e.g., Chronos) in volatile or sparse domains (e.g., power, car parts) while providing more interpretable explanations, whereas foundation models excel only in stable or trend-driven contexts (e.g., finance).