ReasonCast: Towards Explainable Time Series Forecasting with Reasoning
2026-08-03 • Artificial Intelligence
Artificial IntelligenceMachine Learning
AI summaryⓘ
The authors point out that most time series models either focus on making number predictions or explaining data patterns, but not both together. They propose a new model that can predict future values and provide understandable explanations in one combined answer. To test this idea, they created a benchmark called ReasonTS-Bench with common time series patterns and a method called ReasonCast to train large language models for this dual task. Their experiments show that ReasonCast is better at accurate predictions and giving clear, logical reasons than previous models. They also provide their code for others to use.
time seriesforecastinglarge language modelsjoint task learningself-explanationautoregressive modelsbenchmarkfinetuningcausal reasoningReasonTS-Bench
Authors
Seunghan Lee, Jun Seo, Jaehoon Lee, Junhyeok Kang, Sangjun Han, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Soonyoung Lee, Wonbin Ahn
Abstract
Most time series (TS) models are specialized for a single task, either understanding (i.e., returning text answers about a TS) or generation (i.e., returning a numeric forecast). Only recently have unified models begun to handle the two within a single architecture. Even these models, however, produce the two outputs as task-separated paths and cannot predict a series and explain why that prediction arises within a single coherent response. In this paper, we argue for a task-fused model that jointly produces 1) prediction (generation) and 2) selfexplanation (understanding), thereby integrating 1) numerical TS forecasting and 2) interpretable text reasoning within a single response. To enable the systematic study of this capability, we present both a benchmark and a recipe that jointly address the two tasks. The benchmark, ReasonTS-Bench, identifies five fundamental patterns underlying TS and enables the joint evaluation of both tasks. ReasonCast, our recipe for finetuning any LLM to perform both tasks jointly, yields a model that generates a reasoning chain and a forecast together in a single autoregressive pass. Extensive experiments show that ReasonCast outperforms both LLMs and TS models on prediction accuracy while producing verifiable, causal reasoning. Code is available at: https://github.com/seunghan96/reasoncast.