Ze Fu

2026

1. 

   Foci Policy: Focus on Object-Centric Interactions for Relational Manipulation Policies

   Ze Fu, Pinhao Song, Yutong Hu, and 1 more author

   In Conference on Robot Learning, 2026

   Abs arXiv Bib Code Website

   Foci Policy learns relational manipulation from a single demonstration by predicting relative SE(3) trajectories between task-relevant entities. Object-centric spatial and temporal abstractions reduce trajectory variability and isolate task-critical interactions, enabling robust execution across object configurations, scenes, and grippers.

   @inproceedings{fu2026focipolicy,
     title = {Foci Policy: Focus on Object-Centric Interactions for Relational Manipulation Policies},
     author = {Fu, Ze and Song, Pinhao and Hu, Yutong and Detry, Renaud},
     booktitle = {Conference on Robot Learning},
     year = {2026},
     primaryclass = {cs.RO},
   }

2. 

   TASC: Task-Aware Shared Control for Relational Telemanipulation

   Ze Fu, Pinhao Song, Yutong Hu, and 1 more author

   In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2026

   Abs arXiv Bib Code Website

   TASC is a task-aware shared-control framework for teleoperated manipulation. It builds an open-vocabulary interaction graph from visual observations to infer task-level intent, then provides spatially constrained assistance for grasping and object interaction.

   @inproceedings{fu2026tasc,
     title = {TASC: Task-Aware Shared Control for Relational Telemanipulation},
     author = {Fu, Ze and Song, Pinhao and Hu, Yutong and Detry, Renaud},
     booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems},
     year = {2026},
     primaryclass = {cs.RO},
   }

2023

1. 

   TrafficMCTS: A Closed-Loop Traffic Flow Generation Framework with Group-Based Monte Carlo Tree Search

   Ze Fu*, Licheng Wen*, Pinlong Cai, and 3 more authors

   2023

   Abs arXiv Bib Code Video

   Digital twins for intelligent transportation systems are currently attracting great interests, in which generating realistic, diverse, and human-like traffic flow in simulations is a formidable challenge. Current approaches often hinge on predefined driver models, objective optimization, or reliance on pre-recorded driving datasets, imposing limitations on their scalability, versatility, and adaptability. In this paper, we introduce TrafficMCTS, an innovative framework that harnesses the synergy of groupbased Monte Carlo tree search (MCTS) and Social Value Orientation (SVO) to engender a multifaceted traffic flow replete with varying driving styles and cooperative tendencies. Anchored by a closed-loop architecture, our framework enables vehicles to dynamically adapt to their environment in real time, and ensure feasible collision-free trajectories. Through comprehensive comparisons with state-of-the-art methods, we illuminate the advantages of our approach in terms of computational efficiency, planning success rate, intent completion time, and diversity metrics. Besides, we simulate highway and roundabout scenarios to illustrate the effectiveness of the proposed framework and highlight its ability to induce diverse social behaviors within the traffic flow. Finally, we validate the scalability of TrafficMCTS by showcasing its prowess in simultaneously mass vehicles within a sprawling road network, cultivating a landscape of traffic flow that mirrors the intricacies of human behavior.

   @misc{wen2023trafficmcts,
     title = {TrafficMCTS: A Closed-Loop Traffic Flow Generation Framework with Group-Based Monte Carlo Tree Search},
     author = {Fu*, Ze and Wen*, Licheng and Cai, Pinlong and Fu, Daocheng and Mao, Song and Shi, Botian},
     year = {2023},
     primaryclass = {cs.RO},
     video = {https://youtu.be/Tr0QWCasO54}
   }