mlsanityData to 7 Oct 2026

XForecast: Evaluating Natural Language Explanations for Time Series Forecasting

arXiv 2410.14180 ↗First posted 18 Oct 2024cs.CLcited by 4

Not on the map. Why?

Abstract (arXiv)

Time series forecasting aids decision-making, especially for stakeholders who rely on accurate predictions, making it very important to understand and explain these models to ensure informed decisions. Traditional explainable AI (XAI) methods, which underline feature or temporal importance, often require expert knowledge. In contrast, natural language explanations (NLEs) are more accessible to laypeople. However, evaluating forecast NLEs is difficult due to the complex causal relationships in time series data. To address this, we introduce two new performance metrics based on simulatability, assessing how well a human surrogate can predict model forecasts using the explanations. Experiments show these metrics differentiate good from poor explanations and align with human judgments. Utilizing these metrics, we further evaluate the ability of state-of-the-art large language models (LLMs) to generate explanations for time series data, finding that numerical reasoning, rather than model size, is the main factor influencing explanation quality.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 28 of its 42 references that are in the map. First posted on arXiv 18 Oct 2024.

Builds on

28 of 42 references.

References · 28

  1. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  2. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  3. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,400
  4. 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,004
  5. 2023ChatGPT Outperforms Crowd-Workers for Text-Annotation Taskscited by 849
  6. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 737
  7. 2023Can Large Language Models Be an Alternative to Human Evaluations?cited by 729
  8. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 590
  9. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 506
  10. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 491
  11. 2019Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019cited by 270
  12. 2021Monash Time Series Forecasting Archivecited by 256
16 more
  1. 2018Techniques for Interpretable Machine Learningcited by 237
  2. 2020Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?cited by 189
  3. 2023AnnoLLM: Making Large Language Models to Be Better Crowdsourced Annotatorscited by 171
  4. 2023Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanationscited by 85
  5. 2024Language Models Still Struggle to Zero-shot Reason about Time Seriescited by 81
  6. 2021Explaining Time Series Predictions with Dynamic Maskscited by 72
  7. 2021Time Series Prediction using Deep Learning Methods in Healthcarecited by 71
  8. 2023Is GPT-4 a Good Data Analyst?cited by 69
  9. 2018Do Explanations make VQA Models more Predictable to a Human?cited by 51
  10. 2023In Search of Verifiability: Explanations Rarely Enable Complementary Performance in AI-Advised Decision Makingcited by 50
  11. 2024Evaluating Large Language Models on Time Series Feature Understanding: A Comprehensive Taxonomy and Benchmarkcited by 38
  12. 2024Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuningcited by 26
  13. 2024XAI for All: Can Large Language Models Simplify Explainable AI?cited by 21
  14. 2021TCube: Domain-Agnostic Neural Time-series Narrationcited by 5
  15. 2020Series Saliency: Temporal Interpretation for Multivariate Time Series Forecastingcited by 4
  16. 2021LoMEF: A Framework to Produce Local Explanations for Global Model Time Series Forecastscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2026Generating Expressive and Customizable Evals for Timeseries Data Analysis Agents with AgentFuelcited by 1
  2. 2025Interpretable deep convolutional model for nonlinear multivariate time series in complex systemscited by 0
  3. 2025DiffNator: Generating Structured Explanations of Time-Series Differencescited by 0
  4. 2026LineupRL: Verifiable Reinforcement Learning for Time Series Captioning via Caption-to-Series Identificationcited by 0