PeopleofSuperintelligence

Researcher

Mengya Xu

Also known as 徐梦雅

Mengya Xu is a researcher in surgical vision whose work teaches AI to understand and plan robot-assisted operations. After studies at Soochow University and the National University of Singapore, she completed a Ph.D. in biomedical engineering at NUS in 2024 under Hongliang Ren. In 2021, she published a method for generating surgical reports across domain shifts, and in 2024 she published a privacy-preserving continual segmentation framework for robotic surgery in IEEE Transactions on Medical Imaging. In 2025, she introduced Surgical Action Planning with large language models and received the IHU Strasbourg and NDI Bench to Bedside Award at IPCAI. She is a Research Assistant Professor in the Department of Computer Science and Engineering at the Chinese University of Hong Kong.

Timeline

  1. 2014–2019

    Studies at Soochow and NUS

    She studies information engineering at Soochow University from 2014 and completes her B.Eng. in 2018. At the National University of Singapore, she earns an M.Sc. in electrical and computer engineering in 2019, working on biomedical systems and intelligent sensing technologies with Tianyiyi He, Qiongfeng Shi and Chengkuo Lee.

  2. 2019–2024

    PhD at the Medical Mechatronics Lab

    She starts a Ph.D. in biomedical engineering at the National University of Singapore in 2019 and completes it in 2024, in the Medical Mechatronics Lab under Prof. Hongliang Ren. In 2023, she also serves as a research assistant in Yueming Jin's lab at NUS, contributing to research in surgical vision and intelligence.

  3. 2021

    Surgical report generation

    In July 2021, she and co-authors Mobarakol Islam, Chwee Ming Lim and Hongliang Ren post a class-incremental domain adaptation method with a multi-layer transformer that generates surgical reports during robotic surgery. The paper, tested on nephrectomy and transoral robotic surgery datasets, is accepted at MICCAI 2021.

  4. 2024

    Continual segmentation for robotic surgery

    A paper with Mobarakol Islam, Long Bai and Hongliang Ren appears in IEEE Transactions on Medical Imaging in 2024. It presents a privacy-preserving synthetic continual semantic segmentation framework that counters forgetting of earlier instruments without revealing real patient data.

  5. November 2024

    Robot-assisted dissection guidance

    In November 2024, she posts two preprints on robot-assisted endoscopic submucosal dissection (ESD). ETSM pairs dissection trajectory prediction with a confidence-map safety margin for dual-arm robotic ESD and comes with a dataset of 1,849 short clips. PDZSeg segments dissection zones with visual prompts such as scribbles and bounding boxes.

  6. March 2025

    Surgical action planning with LLMs

    In March 2025, she and Zhongzhen Huang, Jie Zhang, Xiaofan Zhang and Qi Dou introduce the Surgical Action Planning task and LLM-SAP, a language-model framework that predicts future actions in robot-assisted surgery from visual inputs. On the CholecT50-SAP dataset, the fine-tuned Qwen2.5-72B-SFT reaches 19.3% higher accuracy than Qwen2.5-72B.

  7. 2025

    A Bench to Bedside Award

    In 2025, she receives the IHU Strasbourg and NDI Bench to Bedside Award at IPCAI.

  8. As of October 2026

    Research Assistant Professor at CUHK

    She is a Research Assistant Professor in the Department of Computer Science and Engineering at The Chinese University of Hong Kong, working in the MedAIR@CUHK Lab. Her research interests cover AI-powered medical data analysis, data-driven medical decision-making with enhanced safety, and MLLM-based contextual intelligence for medical decision support.

Career

  1. NowChinese University of Hong KongResearch Assistant Professor
  2. From 2023National University of SingaporeResearch Assistant

Education

  1. 2024National University of SingaporePh.D.
  2. 2019National University of SingaporeM.Sc. in Electrical and Computer Engineering
  3. 2018Soochow UniversityB.Eng.

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Chinese University of Hong Kong. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. Universities keep AI researchers for a median of 6 years, compared with 2 years at frontier labsIncludes Chinese University of Hong Kong. Across 15,811 dated work stays (6,529 people), median role tenure runs from 6 years at universities to 2 years at frontier labs. We compare all five organization types, starting decades from the 1970s to the 2020s, and the organizations that keep people longest.
  3. The United States employs an estimated 48% of people shaping AIIncludes Chinese University of Hong Kong. We estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Chinese University of Hong Kong. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Mengya Xu yet. Send us a free-licensed photo.

  1. Xu Mengya – CUHK CSE cse.cuhk.edu.hk. Checked 21 October 2026.
  2. Mengya Xu - Homepage xumengyaamy.github.io. Checked 21 October 2026.
  3. Class-Incremental Domain Adaptation with Smoothing and Calibration for Surgical Report Generation arxiv.org. Checked 21 October 2026.
  4. Privacy-Preserving Synthetic Continual Semantic Segmentation for Robotic Surgery arxiv.org. Checked 21 October 2026.
  5. ETSM: Automating Dissection Trajectory Suggestion and Confidence Map-Based Safety Margin Prediction for Robot-assisted Endoscopic Submucosal Dissection arxiv.org. Checked 21 October 2026.
  6. PDZSeg: Adapting the Foundation Model for Dissection Zone Segmentation with Visual Prompts in Robot-assisted Endoscopic Submucosal Dissection arxiv.org. Checked 21 October 2026.
  7. Surgical Action Planning with Large Language Models arxiv.org. Checked 21 October 2026.