ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning

2026-09-30

AI Summary

This study addresses the limitations of isolated modeling, which overlooks shared information, and indiscriminate parameter sharing, which induces negative transfer in multi-property prediction for electrolytes. To overcome these challenges, this work proposes a unified multi-task learning framework based on a Mixture-of-Experts (MoE) architecture. The method leverages directed transfer analysis to guide cross-domain knowledge flow, while an expert pool and router mechanism are designed to independently extract property-specific features and selectively integrate shared knowledge. Furthermore, residual adapters are incorporated to refine feature representations and suppress task interference. Experimental results demonstrate that the proposed model reduces the average prediction error across twelve properties by 14.8% and decreases the conductivity prediction error for sodium electrolytes by 6.7%, thereby achieving precise synergistic multi-property prediction.

Abstract

Electrolyte formulation design requires balancing multiple physicochemical properties, yet existing models often focus on a limited subset. Learning each property in isolation can overlook transferable chemical information, whereas indiscriminate sharing can introduce cross-property interference. Our directed transfer analysis shows that jointly learning two property prediction tasks can improve or degrade prediction relative to separate training, with asymmetric transfer effects between the tasks. We propose ElectrolyteFM, a unified multi-property prediction model which can more accurately predict multiple properties of each electrolyte by effectively identifying and utilizing property-specific features and knowledge shared across properties. More specifically, ElectrolyteFM learns property-specific representations independently and captures cross-property knowledge through a separately trained expert pool. A router selects relevant shared information for each formulation and target property, and property-specific residual adapters convert this information into corrections to the corresponding representation for prediction. Experiments on Electrolyte12 show that ElectrolyteFM reduces normalized mean absolute error averaged across 12 electrolyte properties by 14.8% relative to the strongest electrolyte-specific baseline. On an independent sodium-electrolyte dataset unseen during training, it reduces conductivity mean absolute error by 6.7% relative to the best-performing baseline.

Problem

• Electrolyte property prediction
• Multi-property learning
• Cross-property interference
• Transfer learning

Innovation

• Multi-property prediction
• Cross-property knowledge learning
• Mixture of experts
• Residual adapters
• Electrolyte formulation