WENQIAN (VINCENT)
ZHANG

Robotics & Embodied AI Researcher

I build and study learning systems for real-world robot manipulation—from hardware integration and teleoperation to demonstration data, imitation learning, and VLA evaluation.

Computer Engineering @ UC Riverside · Robotics Intern in San Francisco

View Selected Work View CV

Real-World ManipulationTeleoperationImitation LearningVLA EvaluationHardware Integration

Research & Focus

How can embodied agents infer what matters and act robustly beyond memorized trajectories?

I work across two connected areas: real-world robot learning and multimodal behavior understanding.

My robotics work spans the full stack—building manipulation platforms, integrating cameras and teleoperation systems, collecting demonstrations, training imitation-learning policies, and evaluating vision-language-action models on physical hardware.

At UC Riverside, I built a Franka FR3 robot-learning cell with GELLO teleoperation and evaluated generalist policies under real-world distribution shifts. At an early-stage robotics startup, I am building dual-arm manipulation, data-collection, and autonomous rollout pipelines.

I am also the first author of a paper accepted to the CVPR 2026 CV4Animals Workshop on context-aware multimodal intent inference.

Selected Works

Dual-Arm Manipulation & Imitation Learning

Ongoing

Robotics Startup / ACT / Real-World Data

Built a leader–follower dual-arm platform with multi-camera perception, teleoperation, demonstration collection, ACT policy training, and autonomous rollout tooling at an early-stage robotics startup.

Franka FR3 Robot Learning Cell

↗

Franka / VLA / Real-World Robotics

Built a Franka Research 3 manipulation cell from scratch — bench, real-time control, cameras, gripper, GELLO teleoperation, and in-the-wild VLA rollouts.

CatSignal — CVPR 2026

↗

AI Research / Multimodal Intent Inference

A Bayesian-inspired multimodal framework for inferring intent in non-speaking agents using spatial context, pose dynamics, and audio evidence.

GELLO Build Guide

↗

Franka / Teleoperation / Hardware

A detailed hardware build of a 3D-printed GELLO leader arm for FR3 teleoperation, including Dynamixel bring-up, wiring, spring installation, and calibration pitfalls.

Publication & Recognition

CVPR 2026 CV4Animals Workshop

Context as Prior: Bayesian-Inspired Intent Inference for Non-Speaking Agents with a Household Cat Testbed

First-author accepted poster paper on intent inference for non-speaking agents, using household cats as a real-world testbed for context-aware multimodal reasoning.

Poster AcceptedFirst AuthorarXiv View Project

May 2026 · UC Riverside

Chancellor's Research Fellowship

Undergraduate research fellowship supporting the extension of CatSignal into unsupervised multimodal behavior discovery.

Apr 2026 · UC Riverside

Undergraduate Education Mini-Grant

Travel support to present the accepted paper at the CVPR 2026 CV4Animals Workshop.

Let's Connect

Open to research collaborations and strong opportunities in robotics and embodied AI.

Email LinkedIn GitHub CV

Beyond research: games, photography, travel, quiet time with literature—and Yuanbao, my British Shorthair, who unexpectedly inspired a paper.

WENQIAN (VINCENT) ZHANG