PolyRecommender: A Multimodal Recommendation System for Polymer Discovery

Xin Wang, Yunhao Xiao, Rui Qiao

Keywords: Recommender systems, Multimodal learning, Polymer property prediction, Language models, Graph neural networks

TL;DR: We present a multimodal recommender that leverages chemical language and graph representations in a retrieval-and-ranking pipeline to predict polymer properties and recommend similar candidates for accelerated materials discovery.

Abstract: We introduce PolyRecommender, a multimodal discovery framework that integrates chemical language representations from PolyBERT with molecular graph-based representations from a graph encoder. The system first retrieves candidate polymers using language-based similarity and then ranks them using fused multimodal embeddings according to multiple target properties. By leveraging the complementary knowledge encoded in both modalities, PolyRecommender enables efficient retrieval and robust ranking across related polymer properties. Our work establishes a generalizable multimodal paradigm, advancing AI-guided design for the discovery of next-generation polymers.

Submission Track: Paper Track (Short Paper)

Submission Category: AI-Guided Design

Institution Location: Blacksburg, United States

Submission Number: 140