Reproduce “ILA: A Transferable Interface-Level Adapter for Compliant Humanoid Motion Tracking” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
ZhuoChang Cai, Zihao Yan, Jinxuan Zhu, Yanjie Li, Hao Xiong
The paper introduces ILA, adapting motion references so a pretrained humanoid tracker can yield while retaining responsibility for whole-body execution. It uses observation history to predict arm-reference corrections, leaving tracker architecture and parameters fixed. On a G1 wrist-pull task, each of three ILA-equipped trackers reached the target in all ten trials; this result is limited to that task.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.