UMD Robotics Reading Group

Fall 2026 · University of Maryland, College Park

When	Tuesdays through Oct 6, 5:00–6:00 pm; Fridays from Oct 16, 1:00–2:00 pm
Term	Sep 8 – Dec 18, 2026
Organizer	GAMMA, UMD

About

An informal weekly reading group for graduate students interested in robot learning — manipulation, humanoid control, visuomotor policies, world models, and robotic foundation models. For the Fall 2026 semester, we are planning to focus more on robotic manipulation. Each week one person presents one or two recent papers and we discuss them together. Anyone is welcome to attend, and anyone can sign up to present.

Want to present? Pick an open slot from the schedule below and email Rocktim. Open slots are marked TBD. Note: sessions moved from Tuesdays to Fridays starting Oct 16.

Schedule

Wk	Date	Session
1	Sep 8	Kickoff Reading Group Introduction & Overview Rocktim Jyoti Das · slides
2	Sep 15	No meeting ICRA deadline — no reading group
3	Sep 22	Paper presentation SkildAI S1 & Generalist Gen 1.5 & Behavior Prompting Policy Rocktim Jyoti Das · slides
4	Sep 29	Paper presentation PointZero: 3D Point Track Completion for Learning Transferable 3D Dynamics Hyunjin Kim
5	Oct 6	Paper presentation Forceband Botao He
6	Oct 16	Paper presentation · Fri HumanEgo Leo Wang
7	Oct 23	Paper presentation RealWonder & WonderPlay Wei Liu
8	Oct 30	Paper presentation EyeSight Hand: Design of a Fully-Actuated Dexterous Robot Hand with Integrated Vision-Based Tactile Sensors and Compliant Actuation & DEXOP: A Device for Robotic Transfer of Dexterous Human Manipulation Woo Chul Shin
9	Nov 6	Paper presentation Formal Verification Kristy Sakano
10	Nov 13	No meeting CVPR deadline — no reading group
11	Nov 20	Paper presentation AlphaNIST Kasra Torshizi
12	Nov 27	Paper presentation RoboSimStudio: An Interactive Toolkit for Rapid Real2Sim of Robot Manipulation Tasks Rocktim Jyoti Das
13	Dec 4	Paper presentation Microsaccade-inspired Event Camera for Robotics Botao He
14	Dec 11	Paper presentation Do as I Do: Dexterous Manipulation Data from Everyday Human Videos & What 30,000 Hours of Ego-centric Video Does Not Teach Yohan Abeysinghe
15	Dec 18	Paper presentation Blind Precise Assembly: Proprioceptive Sim-to-Real Peg-in-Hole for Diverse Parts & Sereact FLEX: How Cortex 2.6 Learns from Its Own Experience & Demystifying When and Why VLAs Fail in Contact-Rich Tasks and How to Fix Them & Physical Intelligence: Precise Manipulation with Efficient Online RL & SuperMimic: Scaling Humanoid Whole-Body Control with One Differentiable World Model Rocktim Jyoti Das