PolyRecommender: A Multimodal Recommendation System for Polymer Discovery

TL;DR

PolyRecommender is a multimodal AI system that combines chemical language and molecular graph data to efficiently retrieve and rank polymers for targeted property optimization, advancing polymer discovery.

Contribution

It introduces a novel multimodal framework integrating language and graph representations for polymer discovery, enabling better retrieval and ranking of candidate polymers.

Findings

Effective retrieval of candidate polymers using language similarity.

Robust ranking of polymers based on fused multimodal embeddings.

Advances in AI-guided polymer design.

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.