Zhuoran Li

Undergraduate researcher · Shanghai Jiao Tong University

I am an undergraduate researcher at Shanghai Jiao Tong University, working in IPADS. My research spans embodied AI, robot learning, graph learning, and route planning, with a current focus on reliable vision-language-action policies, agentic manipulation, and benchmarks that make progress measurable.

I am currently a research assistant in Professor Yi Ma's group at HKU; he is also a professor at UC Berkeley. I have been fortunate to work with Professor Jinyu Gu and Professor Haibo Chen at SJTU, Professor Lin Shao at NUS, and Professor Zhengyuan Zhou at NYU. I am grateful for the opportunity to learn from and collaborate with them.

Please feel free to contact me

Program
B.Eng. Software Engineering · 2027
Current appointment
Research Assistant · Yi Ma Group, HKU
Academic standing
GPA 92.5/100 · Rank 3/95

News

Under review

CaTS: Code as Tactile Supervisor was submitted to CVPR 2027; I am the first author.

Award

ManiLadder received the Outstanding Workshop Paper Award at the IROS 2026 Bimanual Robot Learning Workshop, “Scaling vs. Structure?”

Grant

Selected as the student PI of an RMB 100,000 NSFC undergraduate project on OpenHarmony embodied-agent systems.

Paper

RoboHarness was accepted to IROS 2026; I am the first-listed co-first author.

Paper

GKD-Recruiter was accepted as an ICML 2026 regular paper; I am a student co-first author.

Paper

DRNCS appeared in ECML PKDD 2025; I am the sole first author.

Selected work

Research, systems, and benchmarks

CVPR 2027 · Under review · First author

CaTS: Code as Tactile Supervisor

CaTS grounds robot agents in contact-rich manipulation. From a single human RGB-D demonstration and physics-grounded tactile-pattern priors, an agent synthesizes stage-wise tactile supervisors and compiles them into a locally executable program that needs no foundation-model calls at run time. It outperforms all evaluated baselines across 6 real-robot and 20 simulated tasks.

IROS 2026 · First-listed co-first author

RoboHarness: A memory-augmented policy harness for VLA robustness

RoboHarness adapts frozen vision-language-action policies through dual-memory retrieval and attribution-driven interventions. Across three VLA backbones, it improves average success by 56.6 percentage points, including 89.1 points on long-horizon tasks.

arXiv 2026 · IROS 2026 Workshop Outstanding Paper Award · Core developer

ManiLadder: Benchmarking robot manipulation through a categorized, multi-difficulty-level task ladder

What makes a manipulation task easy or hard? ManiLadder answers this systematically: 112 simulated tasks in 4 difficulty levels and 8 task categories, where each level is set by the success rate of an imitation-learning policy under the same training budget. Every level spans grippers and dexterous hands, rigid and deformable objects, and single- and dual-arm settings. Each task ships with demonstrations and trained flow-matching, π0.5 and Astra checkpoints, and the benchmark is not yet solved by Astra.

GKD-Recruiter architecture combining social and worker-task graph embeddings, graph knowledge distillation, and Rainbow DQN recruitment.

ICML 2026 regular paper · Student co-first author

GKD-Recruiter: Graph knowledge distillation for spatial crowdsourcing

GKD-Recruiter combines directional social influence with worker–task affinity and uses graph knowledge distillation plus Rainbow DQN to recruit workers for the right physical tasks under limited demand.

DRNCS dual-level route generation pipeline with shortcut construction and prediction on original and sparse graphs.

ECML PKDD 2025 · First author

DRNCS: Dual-level route generation through contraction and shortcuts

DRNCS builds a sparse road graph with shortcut-edge differential contraction, then generates routes on the contracted structure. It reports 0.3–0.8 ms per trip and more than 40% faster queries across five real-world datasets.

Layered graph for the Shortest Path Tour Problem, showing candidate nodes across ordered stages from source to destination.

Theoretical Computer Science 1054 (2025) · Co-author

Algorithms for the Shortest Path Tour Problem

Two exact, linear-space algorithms solve ordered waypoint tours on road networks. Stage Dijkstra reaches the best-known time bound, while Global Dijkstra is faster in practice on large graphs.

Experience

Research positions and internships

  1. 2026.10Present

    The University of Hong Kong

    Research Assistant

    Advisor
    Yi Ma
    Project
    CaTS (CVPR 2027, under review)
  2. 2026.08Present

    New York University

    Research Intern

    Advisor
    Zhengyuan Zhou
    Project
    DRWM, robust targets for world-model RL (ICML 2027, under review)
  3. 2026.052026.10

    ViTai Robotics

    Intern

    Mentors
    Yizhe Wu and Rui Li
    Focus
    VTLA post-training; agentic robot–tactile integration
  4. 2025.042025.12

    LinS Lab, National University of Singapore

    Research Intern

    Advisor
    Lin Shao
    Project
    ManiLadder (IROS 2026 Workshop, Outstanding Paper Award)
  5. 2024.06Present

    Shanghai Jiao Tong University

    Undergraduate Researcher, IPADS

    Advisors
    Xiaofeng Gao, Jinyu Gu, and Haibo Chen
    Projects
    RoboHarness, GKD-Recruiter, DRNCS, SPTP; Student PI of an NSFC undergraduate grant

Misc

Outside research, I enjoy climbing and strength training, and I like keeping an active routine while exploring new places.