LAMDA Group logo Nanjing University School of Artificial Intelligence logo
Portrait of Jun-Peng Jiang

Jun-Peng Jiang 蒋俊鹏

Ph.D. Candidate

Learning And Mining from DatA (LAMDA) Group
School of Artificial Intelligence, Nanjing University
State Key Laboratory for Novel Software Technology

Email Google Scholar

Short Biography

I received my B.Sc. degree from Soochow University in June 2021. In the same year, I was admitted to pursue a Ph.D. degree at Nanjing University without an entrance examination. I am a member of the LAMDA Group, led by Professor Zhi-Hua Zhou, and am supervised by Professor De-Chuan Zhan and Professor Han-Jia Ye.

Research Interests

My research focuses on building intelligent systems that can transform complex, multimodal information into structured knowledge and actionable insights. In particular, I study how machines can jointly perceive, parse, understand, and reason over tables, documents, text, images, and visual structures, with the broader goal of developing data world models and agents capable of interacting with real-world information.

01

Multimodal Tabular Intelligence

Learning from tables together with text, images, and visual structure to enrich tabular signals.

02

Data World Model

Building data world models/agents that parse, understand, reason over, and act on complex information.

03

Document Intelligence

End-to-end document parsing, OCR, document understanding, and structured information extraction.

Publications

Preprints, conference proceedings, and journal articles.

Preprints

OPD

Dense Supervision, Sparse Updates: On the Sparsity and Geometry of On-Policy Distillation

Guo Yu, Wenlin Liu, Yulan Hu, Hao-Xuan Ma, Jun-Peng Jiang, Han-Jia Ye

Preprint

TABTECO multimodal tabular learning workflow

Twins: Synergizing Tabular Data and Text Signals via Joint Representation Learning

Jun-Peng Jiang, Si-Yang Liu, De-Chuan Zhan, Han-Jia Ye

Preprint

Conference Publications

OvisOCR document parsing benchmark comparison
ICML'26

OvisOCR: End-to-End Document Parsing via Aligning Specialized Perception with General Reasoning

Jun-Peng Jiang, Shiyin Lu, An-Yang Ji, Yinglun Li, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang, De-Chuan Zhan, Han-Jia Ye

Proceedings of the 43rd International Conference on Machine Learning · ICML 2026 · Seoul, South Korea

Comparison of supervised fine-tuning and reinforcement learning for document parsing
ICML'26

Dissecting Post-Training: Uncovering the Complementary Roles of SFT and RL for Document Parsing

Jun-Peng Jiang, An-Yang Ji, Shiyin Lu, Guodong Zheng, Weihong Zhang, Qing-Guo Chen, Weihua Luo, Kaifu Zhang, Long Chen, De-Chuan Zhan, Han-Jia Ye

Proceedings of the 43rd International Conference on Machine Learning · ICML 2026 · Seoul, South Korea

TopBench benchmark for tabular question answering
ICML'26

TopBench: A Benchmark for Implicit Prediction and Reasoning over Tabular Question Answering

An-Yang Ji, Jun-Peng Jiang, De-Chuan Zhan, Han-Jia Ye

Proceedings of the 43rd International Conference on Machine Learning · ICML 2026 · Seoul, South Korea

Multimodal tabular reasoning model overview
NeurIPS'25

Multimodal Tabular Reasoning with Privileged Structured Information

Jun-Peng Jiang, Yu Xia, Hai-Long Sun, Shiyin Lu, Qing-Guo Chen, Weihua Luo, Kaifu Zhang, De-Chuan Zhan, Han-Jia Ye

Advances in Neural Information Processing Systems 38 · NeurIPS 2025

Hawk spatial speculative decoding framework
NeurIPS'25

Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation

Zhi-Kai Chen, Jun-Peng Jiang, Han-Jia Ye, De-Chuan Zhan

Advances in Neural Information Processing Systems 38 · NeurIPS 2025

CoCoTab multimodal tabular understanding method
ICML'25

Compositional Condition Question Answering in Tabular Understanding

Jun-Peng Jiang, Tao Zhou, De-Chuan Zhan, Han-Jia Ye

Proceedings of the 42nd International Conference on Machine Learning · ICML 2025 · Vancouver, Canada

CHARMS knowledge transfer from tables to images
ICML'24

Tabular Insights, Visual Impacts: Transferring Expertise from Tables to Images

Jun-Peng Jiang, Han-Jia Ye, Leye Wang, Yang Yang, Yuan Jiang, De-Chuan Zhan

Proceedings of the 41st International Conference on Machine Learning · ICML 2024 Spotlight · Vienna, Austria

Journal Publications

Overview of tabular representation learning survey
TPAMI'26

Representation Learning for Tabular Data: A Comprehensive Survey

Jun-Peng Jiang, Si-Yang Liu, Hao-Run Cai, Qile Zhou, Han-Jia Ye

IEEE Transactions on Pattern Analysis and Machine Intelligence · TPAMI 2026

TALENT tabular analytics and learning toolbox overview
JMLR'26

TALENT: A Tabular Analytics and Learning Toolbox

Si-Yang Liu, Hao-Run Cai, Qile Zhou, Huai-Hong Yin, Tao Zhou, Jun-Peng Jiang, Han-Jia Ye

Journal of Machine Learning Research · JMLR 2025

CJC overview
CJC'26

Tabular Data Generation for Complex Requirements Based on Large Language Models

Jun-Peng Jiang, Zi-Rong Li, Yi-Chu Xu, Han-Jia Ye. De-Chuan Zhan

Chinese Journal of Computers (计算机学报)· CJC 2026

Honors

Teaching Assistant

Academic Service

01 · Tutorial

Tabular Representation Learning

A tutorial with Prof. Han-Jia Ye at AAAI 2026.

View tutorial

02 · Conference Reviewing

  • ICML2023–2026
  • NeurIPS2023–2025
  • ICLR2023–2025
  • CVPR · KDD · AAAI · IJCAI · AISTATS2024–2026

03 · Journal Reviewing

  • IEEE TPAMI
  • Neural Networks
  • Knowledge-Based Systems

Contact

Email

jiangjp [AT] lamda.nju.edu.cn

Mailing Address

Jun-Peng Jiang
National Key Laboratory for Novel Software Technology, Nanjing University, Xianlin Campus Mailbox 603
163 Xianlin Avenue, Qixia District, Nanjing 210023, China
南京市栖霞区仙林大道163号，南京大学仙林校区603信箱，软件新技术国家重点实验室，210023