3 papers
cs.LG2026
Learning Materials Properties from Scarce Labels and Unlabeled Crystals
Wentao Li, Yizhe Chen, Jiangjie Qiu +3
Learning materials properties from scarce labels and unlabeled crystals is a central challenge for data-driven materials discovery. We present SemiMat, a controlled benchmark for s…
cs.LG2026
CoMPASS: Collaborative Molecular Property Prediction via Adaptive Small-Large Model Synergy
Wentao Li, Jiangjie Qiu, Yijun Li +2
Accurate molecular property prediction requires both statistical reliability and chemical reasoning. Graph neural networks can be calibrated directly on labeled assays but remain l…
cs.AI2026
ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation
Jiangjie Qiu, Yijun Li, Xiaonan Wang
Autonomous chemistry increasingly depends on environments in which agents can repeatedly act, observe, and adapt.Physical laboratories provide essential real-material evidence but…