Beyond Fixed Patches: Enhancing GPTs for Financial Prediction with Adaptive Segmentation and Learnable Wavelets
Not on the map. Why?
Abstract (arXiv)
The extensive adoption of web technologies in the finance and investment sectors has led to an explosion of financial data, which contributes to the complexity of the forecasting task. Traditional machine learning models exhibit limitations in this forecasting task constrained by their restricted model capacity. Recent advances in Generative Pre-trained Transformers (GPTs), with their greatly expanded parameter spaces, demonstrate promising potential for modeling complex dependencies in temporal sequences. However, existing pretraining-based approaches typically focus on fixed-length patch analysis, ignoring market data's multi-scale pattern characteristics. In this study, we propose GPT4FTS, a novel framework that enhances pretrained transformer capabilities for temporal sequence modeling through dynamic patch segmentation and learnable wavelet transform modules. Specifically, we first employ K-means++ clustering based on DTW distance to identify scale-invariant patterns in market data. Building upon pattern recognition results, we introduce adaptive patch segmentation that partitions temporal sequences while preserving pattern integrity. To accommodate time-varying frequency characteristics, we devise a dynamic wavelet transform module that emulates discrete wavelet transformation with enhanced flexibility in capturing time-frequency features. Extensive experiments on real-world financial datasets substantiate the framework's efficacy. The source code is available: hrefhttps://anonymous.4open.science/r/GPT4FTS-6BCC/
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 28 of its 44 references that are in the map. First posted on arXiv 5 May 2025.
Builds on
28 of 44 references.
References · 28
- 2017Attention Is All You Needcited by 60,361
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
- 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
- 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,367
- 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
- 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
- 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
- 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
- 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
- 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
- 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 632
16 more
- 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
- 2023FinGPT: Open-Source Financial Large Language Modelscited by 253
- 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 208
- 2023TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecastingcited by 174
- 2024Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Expertscited by 95
- 2022Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learningcited by 78
- 2023Temporal Data Meets LLM -- Explainable Financial Time Series Forecastingcited by 77
- 2024Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilitiescited by 48
- 2024Position: What Can Large Language Models Tell Us about Time Series Analysiscited by 47
- 2019AlphaStock: A Buying-Winners-and-Selling-Losers Investment Strategy using Interpretable Deep Reinforcement Attention Networkscited by 40
- 2023MASTER: Market-Guided Stock Transformer for Stock Price Forecastingcited by 32
- 2021Fully Learnable Deep Wavelet Transform for Unsupervised Monitoring of High-Frequency Time Seriescited by 29
- 2024Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learningcited by 25
- 2021Deep Graph Convolutional Reinforcement Learning for Financial Portfolio Management -- DeepPocketcited by 13
- 2024StockTime: A Time Series Specialized Large Language Model Architecture for Stock Price Predictioncited by 9
- 2025Learning Universal Multi-level Market Irrationality Factors to Improve Stock Return Forecastingcited by 4
Built on by
Citing papers per year: 2025: 2.
2 papers.