Attention-enhanced variational learning for physically informed discovery of exceptionally hard multicomponent bulk metallic glasses

The discovery of high-performance multicomponent alloys is constrained by the vastness of composition space and the scarcity of experimentally validated data. We develop VIBANN, a variational information bottleneck-augmented attention-based neural network framework, for uncertainty-aware inverse design of exceptionally hard bulk multicomponent metallic glasses. The model learns chemically structured latent representations of alloy composition and indentation load to search the candidate space under constraints of chemical plausibility, novelty, and predictive uncertainty. Guided by this framework, we synthesize five B-Nb-Fe-W-Co/Hf/Ru/Zr-rich bulk multicomponent metallic glasses. All alloys form fully amorphous rods of 2 mm diameter and reach Vickers hardness values of about 2450 HV, among the highest reported for bulk metallic glasses under comparable conditions. Latent space analysis, attribution trends, and molecular dynamics-based atomistic simulations show that exceptional hardness in this compositional space arises from dense atomic packing, boron-enriched short-range environments and refractory-stabilized local rigidity. Together, we show that uncertainty-aware latent space learning can discover bulk metallic glasses that combine amorphous structure with exceptionally high hardness under limited data conditions.