LITE: Modeling Environmental Ecosystems with Multimodal Large Language Models

Haoran Li, Junqi Liu, Zexian Wang, Shiyuan Luo, Xiaowei Jia, Huaxiu Yao

Research Area: Societal implications, LMs on diverse modalities and novel applications

Keywords: Environmental Ecosystems, Foundation Models, Spatio-temporal Data, Multimodal Learning

TL;DR: A novel multimodal large language model for cross-domain environmental ecosystem modeling, demonstrating SOTA performance across all domains and maintaining robustness to incomplete observations and distribution shifts in environmental data.

Abstract: The modeling of environmental ecosystems plays a pivotal role in the sustainable management of our planet. Accurate prediction of key environmental variables over space and time can aid in informed policy and decision-making, thus improving people's livelihood. Recently, deep learning-based methods have shown promise in modeling the spatial-temporal relationships for predicting environmental variables. However, these approaches often fall short in handling incomplete features and distribution shifts, which are commonly observed in environmental data due to the substantial cost of data collection and malfunctions in measuring instruments. To address these issues, we propose LITE -- a multimodal large language model for environmental ecosystems modeling. Specifically, LITE unifies different environmental variables by transforming them into natural language descriptions and line graph images. Then, LITE utilizes unified encoders to capture spatial-temporal dynamics and correlations in different modalities. During this step, the incomplete features are imputed by a sparse Mixture-of-Experts framework, and the distribution shift is handled by incorporating multi-granularity information from past observations. Finally, guided by domain instructions, a language model is employed to fuse the multimodal representations for the prediction. Our experiments demonstrate that LITE significantly enhances performance in environmental spatial-temporal prediction across different domains compared to the best baseline, with a 41.25\% reduction in prediction error. This justifies its effectiveness.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the COLM Code of Ethics on https://colmweb.org/CoE.html

Author Guide: I certify that this submission complies with the submission instructions as described on https://colmweb.org/AuthorGuide.html

Submission Number: 226