Reproduce “TaF-Foundation: Wrench-Grounded Cross-Sensor Tactile Representations” 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
Jiaming Jiang, Yuzhe Huang, Baokun Yang, Hao Liang, Li Gao, Haoran Guo, Pei Lin, Shengcheng Luo and 3 more
TaF-Foundation pairs tactile observations from 24 sensors with simultaneous six-dimensional force-and-torque measurements to train a shared representation without directly matching sensors’ readings. Its dataset contains over 100 million observations across four sensing families. On 21 seen sensors, it improves force prediction and cross-sensor retrieval over AnyTouch 2. Transfer to unseen sensors still requires labeled paired data and sensor-specific adaptation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.