UniPolymer: A Unified Framework for Property Prediction, Structure Recommendation, and Evaluation in Polyimide Design

2026-07-31

AI Summary

This work addresses the lack of effective consistency evaluation between generated structures and target glass transition temperatures (Tg) in traditional polyimide design, which often leads to numerous low-quality candidates advancing to experimental validation. The authors propose the first end-to-end unified framework that integrates property prediction, target-guided generation, and consistency assessment. By leveraging self-supervised chemical semantic learning and multi-scale information fusion, the framework establishes an accurate structure–property mapping. A continuous–discrete joint Tg representation drives autoregressive SELFIES-based molecular generation, while a frozen predictor combined with polyimide-specific structural constraints filters high-consistency candidates. Experiments demonstrate a property prediction R² of 0.93 and a candidate pass rate of 73.79%, substantially outperforming existing methods. Moreover, the predicted Tg values of recommended structures show excellent agreement with molecular dynamics simulations, significantly reducing futile experimental efforts.

Abstract

Designing polyimide structures with specific glass transition temperatures (Tg) is highly challenging. Existing methods primarily focus on target-conditioned generation, lacking an assessment of the consistency between the generated structure and the target properties. This leads to low-quality candidates deviating from the design objective entering subsequent processes, increasing invalid experiments and prolonging the development cycle. To address this issue, we propose UniPolymer, a unified framework for property prediction, target-conditioned generation, candidate evaluation, and structure recommendation in polyimide design and a dataset containing 10066 deduplicated polyimide repeating units with Tg tags (PITg-Curated) was constructed. To improve the consistency between generated candidate structures and the target Tg, UniPolymer first establishes a reliable structure-property relationship mapping through self-supervised chemical semantic learning, structural consistency enhancement, and multi-scale information fusion. Subsequently, the model employs a continuous-discrete joint Tg representation to guide the autoregressive generation of SELFIES. The generated candidate structures are further evaluated using a frozen property predictor and polyimide-specific structural constraints, and ranked according to their deviation from the target Tg, thereby preventing structures deviating from the target from entering the subsequent validation stage. Experimental results show that UniPolymer achieved a property prediction accuracy of R^2=0.93 and a candidate structure evaluation pass rate of 73.79%, which are 2% and 1.21% higher than the best baseline, respectively. Meanwhile, the predicted Tg values of the recommended candidates are in high agreement with the results of molecular dynamics simulations, thereby reducing the number of candidates that enter the high-cost experimental stage.

Problem

• polyimide design
• glass transition temperature
• structure-property consistency
• candidate evaluation
• target-conditioned generation

Innovation

• polyimide design
• structure-property relationship
• target-conditioned generation
• SELFIES
• candidate evaluation