# Curse of Attention: A Kernel-Based Perspective for Why Transformers Fail to Generalize on Time Series Forecasting and Beyond

arXiv: https://arxiv.org/abs/2412.06061 · HTML: https://mlsanity.com/paper/2412.06061/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2024.

## Built on by (most cited first, 8 of 8)

- [The Computational Limits of State-Space Models and Mamba via the Lens of Circuit Complexity](https://mlsanity.com/paper/2412.06148/) (2024)
- [Theoretical Constraints on the Expressive Power of mathsfRoPE-based Tensor Attention Transformers](https://mlsanity.com/paper/2412.18040/) (2024)
- [Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation](https://mlsanity.com/paper/2502.00500/) (2025)
- [Circuit Complexity Bounds for Visual Autoregressive Model](https://mlsanity.com/paper/2501.04299/) (2025)
- [Theoretical Guarantees for High Order Trajectory Refinement in Generative Flows](https://mlsanity.com/paper/2503.09069/) (2025)
- [T2VTextBench: A Human Evaluation Benchmark for Textual Control in Video Generation Models](https://mlsanity.com/paper/2505.04946/) (2025)
- [Provable Failure of Language Models in Learning Majority Boolean Logic via Gradient Descent](https://mlsanity.com/paper/2504.04702/) (2025)
- [Time Series Forecasting Through the Lens of Dynamics](https://mlsanity.com/paper/2507.15774/) (2025)

## Abstract (arXiv)

The application of transformer-based models on time series forecasting (TSF) tasks has long been popular to study. However, many of these works fail to beat the simple linear residual model, and the theoretical understanding of this issue is still limited. In this work, we propose the first theoretical explanation of the inefficiency of transformers on TSF tasks. We attribute the mechanism behind it to bf Asymmetric Learning in training attention networks. When the sign of the previous step is inconsistent with the sign of the current step in the next-step-prediction time series, attention fails to learn the residual features. This makes it difficult to generalize on out-of-distribution (OOD) data, especially on the sign-inconsistent next-step-prediction data, with the same representation pattern, whereas a linear residual network could easily accomplish it. We hope our theoretical insights provide important necessary conditions for designing the expressive and efficient transformer-based architecture for practitioners.