Papers
10
Total Citations
225
H-Index
8
About
Yuhong Deng is a leading researcher in robotic manipulation, with a focus on enabling robots to interact with complex, deformable objects in unstructured environments. Her work spans deep reinforcement learning, interactive perception, and language-conditioned manipulation, with a particular emphasis on grasping, cloth folding, and deformable object rearrangement. Deng’s most cited paper (87 citations) introduces a novel deep reinforcement learning framework for robotic pushing and picking in cluttered environments, featuring a composite suction-gripper hand for stable grasping. She also pioneered the concept of Manipulation Question Answering (MQA), where robots physically alter their environment to answer questions—a creative bridge between robotics and AI reasoning. Her recent work on Foldsformer (2022) applies space-time attention to multi-step cloth manipulation, while her 2024 study on language-conditioned deformable object manipulation uses graph dynamics to enable multi-task learning. With over 225 total citations and a growing portfolio of high-impact publications, Deng is recognized for advancing the frontier of deformable object manipulation, making household robots more capable of handling tasks like folding clothes and rearranging objects in smart city warehouses.
Research Focus
Key Achievements
Top Papers
- 1
- 2An Interactive Perception Method for Warehouse Automation in Smart Cities32 citations · 2020
- 3
- 4MQA: Answering the Question via Robotic Manipulation21 citations · 2021
- 5
- 6
- 7Active Affordance Exploration for Robot Grasping13 citations · 2019
- 8
- 9
- 10General-Purpose Clothes Manipulation with Semantic Keypoints5 citations · 2025