David Held
Associate Professor
Studies robot learning for perception and manipulation, including deformable objects.
David Held's lab at CMU works on manipulation & robot learning, robot perception & SLAM and physics-based simulation & animation. 39 papers since 2023, 30 at top venues, mostly CoRL and RSS. 3 PhD students and 12 master's students. 60 alumni, most often to Amazon and Adobe. Applications close .
At a glance
- papers, 2023–2026
- 10/ year
- at T1 venues
- 77%
- PhD students
- 3
- MS students
- 12
- alumni listed
- 60
Papers by yearfrom the lab homepage, 2023–2026
- 8 papers2026 · T1 8 · collecting
- 6 papers2025 · T1 6
- 15 papers2024 · T1 10
- 10 papers2023 · T1 6
T1 venuesOtherVenues: CoRL, RSS, ICML, ICRA, IROS
Research focuspaper titles that use each topic's words
- Manipulation & robot learning25 papers
- Robot perception & SLAM7 papers
- Physics-based simulation & animation2 papers
- deformable objects1 paper
- deformable1 paper
Paper list
From the lab homepage.
20268 papers
- ICML (T1)3D-DLP: Self-supervised 3D Object-centric Scene Representation Learning
- ICRA (T1)Disentangled Point Diffusion for Precise Object Placement (opens in a new tab)
- CoRL (T1)From Grasps to Dexterity: Large-Scale Grasp Pretraining for Dexterous Manipulation
- RSS (T1)GHOST: Hierarchical Sub-Goal Policies for Generalizing Robot Manipulation
- CoRL (T1)H2RBench: A Real-to-Sim Benchmark for Evaluating Human-to-Robot Transfer
- ICLR (T1)Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling (opens in a new tab)
- CoRL (T1)Long-Horizon Multi-Object Rearrangement via Distilled Tree Search over Continuous Actions
- CoRL (T1)Scouting the Dynamics Gap: Test-Time Policy Adaptation via Action-Outcome Feedback
20256 papers
- RSS (T1)ArticuBot: Learning Universal Articulated Object Manipulation Policy via Large Scale Simulation
- CoRL (T1)Force-Modulated Visual Policy for Robot-Assisted Dressing with Arm Motions
- CoRL (T1)Learn from What We HAVE: History-Aware VErifier that Reasons about Past Interactions Online
- CoRL (T1)Planning from Point Clouds over Continuous Actions for Multi-object Rearrangement
- IROS (T1)Real-World Offline Reinforcement Learning from Vision Language Model Feedback
- ICRA (T1)SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies Using Gaussian Splatting
202415 papers
- Autonomous RobotsActive Velocity Estimation using Light Curtains via Self-Supervised Multi-Armed Bandits
- ICLR (T1)Deep SE(3)-Equivariant Geometric Reasoning for Precise Placement Tasks (opens in a new tab)
- NeurIPS (T1)DiffTOP: Differentiable Trajectory Optimization for Deep Reinforcement and Imitation Learning
- CoRL (T1)FlowBotHD: History-Aware Diffuser Handling Ambiguities in Articulated Objects Manipulation
- RA-LForce-Constrained Visual Policy: Safe Robot-Assisted Dressing via Multi-Modal Sensing
- RSS (T1)HACMan++: Spatially-Grounded Motion Primitives for Manipulation
- IEEE International Conference on Robotics and Automation (ICRA)Learning Distributional Demonstration Spaces for Task-Specific Cross-Pose Estimation
- IROS (T1)Learning Generalizable Tool-use Skills through Trajectory Generation
- CoRL (T1)Modeling Drivers' Situational Awareness from Eye Gaze for Driving Assistance
- CoRL (T1)Non-rigid Relative Placement through 3D Dense Diffusion
- RA-LObject Importance Estimation using Counterfactual Reasoning for Intelligent Driving
- ICML (T1)RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedback
202310 papers
- IEEE International Conference on Robotics and Automation (ICRA)AutoBag: Learning to Open Plastic Bags and Insert Objects
- IROS (T1)Bagging by Learning to Singulate Layers Using Interactive Perception
- IEEE International Conference on Robotics and Automation (ICRA)EDO-Net: Learning Elastic Properties of Deformable Objects from Graph Dynamics
- IEEE International Conference on Robotics and Automation (ICRA)Elastic Context: Encoding Elasticity for Data-driven Models of Textiles
- CoRL (T1)FlowBot++: Learning Generalized Articulated Objects Manipulation via Articulation Projection
- CoRL (T1)HACMan: Learning Hybrid Actor-Critic Maps for 6D Non-Prehensile Manipulation
- ICRA (T1)Neural Grasp Distance Fields for Robot Manipulation
- RSS (T1)One Policy to Dress Them All: Learning to Dress People with Diverse Poses and Garments
- CVPR (T1)Point Cloud Forecasting as a Proxy for 4D Occupancy Forecasting
- IEEE International Conference on Robotics and Automation (ICRA)Self-supervised Cloth Reconstruction via Action-conditioned Cloth Tracking
Where alumni went60 listed, 30% to industry
Industry 12Academia 28
Most often to
- Amazon
- Adobe
- Apple
Also to
- Carnegie Mellon University
- UT Austin
- Boston Dynamics AI Institute -> AI Institute
Apply to CMUFall 2027
CMU SCS Computer Science PhD
- Round
- Single round
- Opens
- GRE
- See program page
- English
- TOEFL / IELTS / Duolingo (F-1/J-1 non-native speakers)
- Fee
- USD 100 (USD 80 by early deadline)
Confirmed on the program page.
Similar labsnearest by research summary
- USUCLA
Jiaqi Ma
- Planning, control & multi-robot systems
- USGeorgia Tech
Patricio Antonio Vela
- Robot perception & SLAM
- USUMD
Ming C. Lin
- Physics-based simulation & animation
- CNPeking
Mengyuan Liu
- Robot perception & SLAM
- CNZhejiang
Yue Wang
- Manipulation & robot learning
- CNSJTU
Ling Pei
- Robot perception & SLAM
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
CMUcloses