Yifeng Cao (曹逸峰)

Yifeng Cao (曹逸峰)

he/his

Columbia University

Publications

You can also find my articles on my Google Scholar profile.

Conference Papers

Uncertainty Comes for Free: Human-in-the-Loop Policies with Diffusion Models

Published in Feb 26, 2025

We leverage the generative process of diffusion policies to compute an uncertainty-based metric based on which the autonomous agent can decide to request operator assistance at deployment time, without requiring any operator interaction during training. Additionally, we show that the same method can be used for efficient data collection for fine-tuning diffusion policies in order to improve their autonomous performance.

Recommended citation: He, Zhanpeng, Yifeng Cao, and Matei Ciocarlie. "Uncertainty Comes for Free: Human-in-the-Loop Policies with Diffusion Models." arXiv preprint arXiv:2503.01876 (2025).
Download Paper

Learning Precise, Contact-Rich Manipulation through Uncalibrated Tactile Skins

Published in Oct 22, 2024

ViSk builds on BAKU, a SOTA policy architecture, and AnySkin – a magnetic tactile sensor. We present comprehensive evaluations on four tasks requiring mm-scale precision: Plug insertion, USB insertion, Card swiping, and Book retrieval, and see an average improvement of ~27.5% when using ViSk over vision-only policies across the four tasks. Additionally, the most exciting part of ViSk is the extent of generalizability of learned policies; it can also perform really well in both unseen spatial configurations of the environment as well as unseen variants of the grasped objects.

Recommended citation: Pattabiraman, Venkatesh, Yifeng Cao, Siddhant Haldar, Lerrel Pinto, and Raunaq Bhirangi. "Learning Precise, Contact-Rich Manipulation through Uncalibrated Tactile Skins." arXiv preprint arXiv:2410.17246 (2024).
Download Paper

AnySkin: Plug-and-play Skin Sensing for Robotic Touch

Published in Sep 27, 2024

While tactile sensing is widely accepted as an important and useful sensing modality, its use pales in comparison to other sensory modalities like vision and proprioception. AnySkin addresses the critical challenges that impede the use of tactile sensing – versatility, replaceability, and data reusability.

Recommended citation: Bhirangi, Raunaq, Venkatesh Pattabiraman, Enes Erciyes, Yifeng Cao, Tess Hellebrekers, and Lerrel Pinto. "AnySkin: Plug-and-play Skin Sensing for Robotic Touch." arXiv preprint arXiv:2409.08276 (2024).
Download Paper