News and insights

Research, collaboration and company updates from SparseMind.

COMPANYNeurIPS 2026

Two poster papers explore hardware-friendly semi-structured sparse training and channel-wise dynamic sparsity for convolutional networks.

PARTNERSHIPS2026.10.21

CHT and Moffett achieve up to 2.89× execution speedup in a same-platform compiler evaluation of a 7B model.

INSIGHTSSPARSEMIND PERSPECTIVE

Studying network connectivity and efficiency across models and hardware.

COMPANYNSIA 2026

Supporting research exchange in network science and AI.

Contact us