Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecasting

TL;DR

Time-VLM introduces a multimodal framework utilizing vision and language models to enhance time series forecasting by capturing temporal, visual, and textual information, especially effective in few-shot and zero-shot scenarios.

Contribution

The paper presents a novel multimodal approach that combines pre-trained vision-language models with time series data for improved forecasting accuracy.

Findings

Outperforms existing models in few-shot scenarios

Effective in zero-shot time series forecasting

Demonstrates the benefit of multimodal fusion for temporal data

Abstract

Recent advancements in time series forecasting have explored augmenting models with text or vision modalities to improve accuracy. While text provides contextual understanding, it often lacks fine-grained temporal details. Conversely, vision captures intricate temporal patterns but lacks semantic context, limiting the complementary potential of these modalities. To address this, we propose \method, a novel multimodal framework that leverages pre-trained Vision-Language Models (VLMs) to bridge temporal, visual, and textual modalities for enhanced forecasting. Our framework comprises three key components: (1) a Retrieval-Augmented Learner, which extracts enriched temporal features through memory bank interactions; (2) a Vision-Augmented Learner, which encodes time series as informative images; and (3) a Text-Augmented Learner, which generates contextual textual descriptions. These components collaborate with frozen pre-trained VLMs to produce multimodal embeddings, which are then fused with temporal features for final prediction. Extensive experiments demonstrate that Time-VLM achieves superior performance, particularly in few-shot and zero-shot scenarios, thereby establishing a new direction for multimodal time series forecasting. Code is available at https://github.com/CityMind-Lab/ICML25-TimeVLM.