Xiao Ma (马骁)

I am a research scientist at TikTok, Singapore. Prior to that, I pursued my PhD at National University of Singapore, advised by Prof. David Hsu. I have worked as research scientist at Dyson Robot Learning Lab, led by Dr. Stephen James, at Sea AI Lab, hosted by Prof. Shuicheng Yan and Dr. Min Lin, and at SenseTime Research, hosted by Dr. Shuai Yi.

My research spans across general decision making, robotics, and multi-modal models. My long-term goal is to develop intelligent embodied agents that perceive, interact with, and adapt to unstructured physical environments in a shared autonomy with humans.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

photo credits to Rachel Li Rui ♥

- [Jan. 2025] 1 paper accepted to ICRA 2025.

- [Sept. 2024] 1 paper accepted to CoRL 2024.

- [Jul. 2024] I joined ByteDance Research as a research scientist.

- [Jun. 2024] 1 paper accepted to RA-L 2024.

- [Feb. 2024] 1 paper accepted to CVPR 2024.

- [Sept. 2023] 3 papers accepted to NeurIPS 2023.

- [Feb. 2023] 1 paper accepted to CVPR 2023.

- [Jan. 2023] 3 papers accepted to ICLR 2023 (1 oral 2 posters)!

- [Feb. 2022] G-DOOM for deformable object manipulation has been accepted to ICRA 2022!

- [May 2021] PROMPT for ab-initio object manipulation has been accepted by RSS 2021.

- [Oct. 2020] CVRL for model-based RL under complex observations has been accepted by CoRL 2020.

- [Sept. 2020] BALMS for long-tailed visual recognition has been accepted by NeurIPS 2020.

- [Jul. 2020] STAR for pedestrian trajectory prediction has been accepted by ECCV 2020.

- [Dec. 2019] DPFRL for reinforcement learning under complex and partial observations has been accepted by ICLR 2020.

- [Nov. 2019] PF-RNNs for sequence modeling under uncertainty has been accepted to AAAI 2020.

- [Jun. 2019] DAN was nominated for the best system paper and best student paper of RSS 2019!

RoboVLM is a unified, open-source and flexible VLA framework for easy integration of any VLMs for robotic tasks, within just 30 lines of code!

We present BiGym, a new benchmark and learning environment for mobile bi-manual demo-driven robotic manipulation. BiGym consists of 40 diverse tasks in home environment, and provides human-collected demonstrations.

GreenAug provides a simple visual augmentation to robot policies by first collecting data with a green screen, then augmenting it with different textures. The resulting policy can be transferred to unseen visually distinct novel locations (scenes).

We present a new family of action spaces that benefits from both the efficiency from the task space and the flexibility from the joint space for the robotic manipulation.

Hierarchical Diffusion Policy (HDP) factorises the policy space into a 1) high-level task-planning agent and 2) low-level goal-conditioned diffusion policy, which achieves both task-level generalisation and flexible low-level control.

InsActor is a principled generative framework that leverages recent advancements in diffusion-based human motion models for physics-based human animation generation.

We introduce Efficient Diffusion Policies (EDPs), a more general, faster, and better diffusion policy class for offline RL. EDPs reduce the training time of DQL from 5 days to 5 hours!

MISA is a general framework for offline RL motivated by mutual information estimation. We show that both Conservative Q Learning (CQL) and TD3+BC can be considered as its variants.

We present Imitation Learning via Differentiable Physics (ILD), which casts the imitation learning as a state-matching task through differentiable physics-based Chamfer distance loss. ILD significantly improves the sample efficiency and generalization of imitation learning algorithms with only one expert demonstration.

We present DaXBench, a comprehensive benchmark for deformable object manipulation, including planning, imitation learning, and reinforcement learning, based on a scalable and differentiable physics simulator coded in JAX.

DiffMimic scales motion imitation for simulated characters with differentiable physics. Training controllers on large-scale motion database is more accessible with DiffMimic.

We present G-DOOM for deformable object manipulation. G-DOOM abstract an deformable object as a keypoint-based graph and models the spatio-temporal keypoint interactions with Recurrent Graph Dynamics. G-DOOM achieves SOTA performance on a set of deformable object manipulation tasks.

This paper introduces PROMPT, a framework for particle-based object manipulation. PROMPT performs high-quality online point cloud reconstruction from multi-view images captured by an eye-in-hand camera. It achieves high performance in object grasping, pushing, and placing.

Our key observation is that softmax is biased under the long-tailed distribution. BALMS provides a mathematically unbiased gradient estimate for long-tailed distributions and applies meta-learning to further improve the data sampling process.

We introduce STAR, the first transformer-based pedestrian trajectory predictor. STAR generalizes the Transformers into spatio-temporal graphs and significantly improves the trajectory prediction accuracy (2x).

We introduce PF-RNNs for general sequence prediction under uncertainty. PF-RNNs encodes a differentiable particle filter algorithm with standard RNNs and improves the general sequence prediction performance.

A DAN is composed of neural network modules, each encoding a differentiable algorithm and an associated model; and it is trained end-to-end from data. The algorithms and models act as structural priors to reduce the data requirements for learning; end-to-end learning allows the modules to adapt to one another and compensate for imperfect models and algorithms.

Stolen from Jon Barron