ProvMind: Provenance-grounded reasoning for materials synthesis

2026-05-27

AI Summary

This work addresses the challenge of optimizing materials synthesis processes, which entails modeling complex causal dependencies and provenance structures often oversimplified as linear sequences in existing approaches. To overcome this limitation, the authors propose ProvMind, a novel framework that constructs MatPROV—a literature-derived provenance graph capturing intricate synthesis relationships—and introduces MatProcBench, the first provenance-driven benchmark for materials synthesis reasoning with dual out-of-distribution (OOD) evaluation settings. ProvMind leverages retrieval of analogous synthesis procedures, generates provenance-aware option compatibility scores, and integrates large language models for constrained reasoning. Under rigorous dual-OOD splits, ProvMind achieves 52.84% accuracy, substantially outperforming baseline methods based on prompt engineering, retrieval-augmented generation, and supervised fine-tuning.

Abstract

Materials process optimization requires reasoning over routes, conditions, tools and causal dependencies, yet most computational formulations flatten synthesis procedures into text or ordered steps. We introduce MatProcBench, a provenance-grounded benchmark constructed from literature-mined MatPROV graphs, to evaluate seven process-reasoning tasks spanning route continuity, step-level variable inference and global causal consistency under both same-split and shift-aware evaluation, including a strict dual-OOD split that combines temporal and material-class shift. We further introduce ProvMind, a process-memory reasoning framework that retrieves analogous training processes, converts them into provenance-aware option-level compatibility scores, and uses a language model for constrained final decision making. ProvMind achieves 52.84\% accuracy on the dual-OOD split, outperforming prompting, retrieval-augmented and supervised fine-tuning baselines.

Problem

• materials synthesis
• process reasoning
• provenance
• causal dependencies
• out-of-distribution generalization

Innovation

• provenance-aware reasoning
• materials synthesis
• process-memory framework
• dual-OOD evaluation
• MatProcBench