LLMForecaster: Improving Seasonal Event Forecasts with Unstructured Textual Data
This addresses the issue of inaccurate demand forecasts for retailers during seasonal events, though it is incremental as it builds on existing forecasting pipelines.
The paper tackled the problem of time-series forecasting models failing to utilize unstructured textual data, leading to inaccurate predictions for seasonal events like holidays, and introduced LLMForecaster, a fine-tuned large language model that improved forecasts in a retail application with statistically significant gains.
Modern time-series forecasting models often fail to make full use of rich unstructured information about the time series themselves. This lack of proper conditioning can lead to obvious model failures; for example, models may be unaware of the details of a particular product, and hence fail to anticipate seasonal surges in customer demand in the lead up to major exogenous events like holidays for clearly relevant products. To address this shortcoming, this paper introduces a novel forecast post-processor -- which we call LLMForecaster -- that fine-tunes large language models (LLMs) to incorporate unstructured semantic and contextual information and historical data to improve the forecasts from an existing demand forecasting pipeline. In an industry-scale retail application, we demonstrate that our technique yields statistically significantly forecast improvements across several sets of products subject to holiday-driven demand surges.