Article

MambaAct: Reducing Inference Latency of Flow-Matching VLAs with a Compact Mamba Action Expert

MambaAct proposes replacing the Transformer-based action expert in π0.5 with a compact Mamba action expert while retaining its pretrained vision-language prefix encoder.

Part of the event: MambaAct uses a compact Mamba action expert to reduce flow-matching VLA inference latency

This page provides a summary and source link only.