Reproduce “GestAdapt: Workspace-Conditioned Co-Speech Gesture Generation for Humanoid Robots” 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
Bosong Ding, Xianglin Zhang, Miao Xin, Murat Kirtay, Giacomo Spigler
GestAdapt generates robot co-speech gestures conditioned on a wrist workspace. Speech and workspace boundaries condition a diffusion Transformer that produces motion. In Reachy2 rankings, GestAdapt placed first in 69.7% of comparisons, versus 19.7% for the no-workspace variant and 10.6% for reference motions. All alternatives shared downstream constraints; this measures preference, not a safety guarantee.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.