Fangkai JIAO

Education

Nanyang Technological University & Infocomm Research (I2R), A*STAR 2022.8 - 2026.6 (expected)

Shandong University 2019.9 - 2022.6

GPA: 90.12/100

Shandong University 2015.9 - 2019.6

GPA: 85.03/100

Research Interests

Publications

* means equal contribution.

Self/Weak-Supervised Learning & Machine Reasoning & LLM

Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks

Xingxuan Li, Weiwen Xu, Ruochen Zhao, Fangkai Jiao, Shafiq Joty, Lidong Bing. Under Review by ICLR.
[Paper]

Preference Optimization for Reasoning with Pseudo Feedback

Fangkai Jiao, Geyang Guo, Xingxing Zhang, Nancy F. Chen, Shafiq Joty, Furu Wei. Under Review by ICLR.

Describe-then-Reason: Improving Multimodal Mathematical Reasoning through Visual Comprehension Training

Mengzhao Jia, Zhihan Zhang, Wenhao Yu, Fangkai Jiao, Meng Jiang. Under Review by ARR.
[Paper]

Relevant or Random: Can LLMs Truly Perform Analogical Reasoning?

Chengwei Qin*, Wenhan Xia*, Tan Wang*, Fangkai Jiao, Yuchen Hu, Bosheng Ding, Ruirui Chen, Shafiq Joty. Under Review by ARR.
[Paper]

How Much are LLMs Contaminated? A Comprehensive Survey and the LLMSanitize Library

Mathieu Ravaut*, Bosheng Ding*, Fangkai Jiao, Hailin Chen, Xingxuan Li, Ruochen Zhao, Chengwei Qin, Caiming Xiong, Shafiq Joty. Preprint.
[Paper][Tool]

Learning Planning-based Reasoning by Trajectories Collection and Process Reward Synthesizing

Fangkai Jiao, Chengwei Qin, Zhengyuan Liu, Nancy F. Chen, Shafiq Joty. EMNLP 2024.
[Paper][Code & Data & Weights]

Improving In-context Learning via Bidirectional Alignment

Chengwei Qin, Wenhan Xia, Fangkai Jiao, Shafiq Joty. Under Review by NeurIPS.
[Paper]

ChatGPT’s One-year Anniversary: Are Open-Source Large Language Models Catching up?

Hailin Chen*, Fangkai Jiao*, Xingxuan Li*, Chengwei Qin*, Mathieu Ravaut*, Ruochen Zhao*, Caiming Xiong, Shafiq Joty. Under Review by COLM.
[Paper][Repo]

UNK-VQA: A Dataset and A Probe into Multi-modal Large Models’ Abstention Ability

Yanyang Guo, Fangkai Jiao, Zhiqi Shen, Liqiang Nie, Mohan Kankanhalli. TPAMI.
[Paper]

SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoning

Bin Wang*, Zhengyuan Liu*, Xin Huang, Fangkai Jiao, Yang Ding, Ai Ti Aw, Nancy F. Chen. NAACL 2024.
[Paper][Data][Leaderboard]

LogicLLM: Exploring Self-supervised Logic-enhanced Training for Large Language Models

Fangkai Jiao, Zhiyang Teng, Bosheng Ding, Zhengyuan Liu, Nancy F. Chen, Shafiq Joty. NAACL 2024.
[Paper] [Code]

MERIt: Meta-Path Guided Contrastive Learning for Logical Reasoning

Fangkai Jiao, Yangyang Guo, Xuemeng Song, Liqiang Nie. Findings of ACL 2022.
[Paper] [Code]

REPT: Bridging Language Models and Machine Reading Comprehension via Retrieval-Based Pre-training

Fangkai Jiao, Yangyang Guo, Yilin Niu, Feng Ji, Feng-Lin Li, Liqiang Nie. Findings of ACL 2021.
[Paper] [Code]

A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction

Yilin Niu*, Fangkai Jiao*, Mantong Zhou, Ting Yao, Jingfang Xu and Minlie Huang. ACL 2020.
[Paper] [Code]

Multimodal Information Retrieval

Retrieving Multimodal Information for Augmented Generation: A Survey

Ruochen Zhao, Hailin Chen, Weishi Wang, Fangkai Jiao, Xuan Long Do, Chengwei Qin, Bosheng Ding, Xiaobao Guo, Minzhi Li, Xingxuan Li, Shafiq Joty. Findings of EMNLP 2023.
[Paper]

Enhanced Multi-domain Dialogue State Tracker with Second-order Slot Interactions

Fangkai Jiao, Yangyang Guo, Minlie Huang, and Liqiang Nie. TASLP 2022.
[Paper] [Code]

Personalized Fashion Compatibility Modeling via Metapath-guided Heterogeneous Graph Learning

Weili Guan, Fangkai Jiao, Xuemeng Song, Haokun Wen, Chung-Hsing Yeh and Xiaojun Chang. SIGIR 2022.
[Paper] [Code]

Liqiang Nie, Fangkai Jiao, Wenjie Wang, Yinglong Wang, and Qi Tian. Transactions on Image Processing (TIP) 2021.
[Paper] [Code]

Projects

Learning from Pseudo Feedback for Code Generation and Reasoning

Done during internship at MSRA.

PandaLLM

Chinese Large Language Models based on LLaMA & Llama2. [code][technical report] (1k stars)

llama-pipeline-parallel

A prototype repository for hybrid training of pipeline parallel and distributed data parallel based on DeepSpeed. [code]

dpo-trajectory-reasoning

An unified code base for DPO/PPO/GRPO training with support to tensor parallel and process reward model. [code]

Experience

Microsoft Research Asia 2024.5 - 2024.11

Research Intern. Advised by Dr. Xingxing Zhang.
Working on weak-supervised learning for machine reasoning and code generation based on LLMs.

Langboat Technology 2022.4 - 2022.7

Intern. Mentored by Yulong Wang and advised by Dr. Ming Zhou
Working on semantic parsing (NL2SQL) and symbolic machine reasoning.

ByteCamp, Bytedance 2021.8.1 - 2021.8.7

Working on self-supervised pre-training for dialogue response selection.

Damo Academy, Alibaba Group 2020.7 - 2021.2

Research Intern. Advised by Dr. Feng Ji and Dr. Feng-Lin Li
Working on Pre-training for complex reading comprehension.

CoAI Group, Tsinghua University 2018.10 - 2019.8

Research Intern. Advised by Prof. Minlie Huang
Working on weak-supervised reading comprehension and dialogue state tracking.

Honors & Awards

Singapore International Graduate Award (SINGA) Agency for Science, Technology & Research (A*STAR), 2022
Dean’s Scholarship School of Computer Science and Technology, Shandong University, 2021
Dean Scholarship School of Software, Shandong University, 2018
Bronze Medal in the ACM-ICPC Asia Regional Contest Urumqi Site ICPC, 2017
Silver Medal in the ACM-ICPC Asia Regional Contest Qingdao Site ICPC, 2017
Bronze Medal in the ACM-ICPC Asia Regional Contest China-Final ICPC, 2016
Silver Medal in the ACM-ICPC Asia Regional Contest Qingdao Site ICPC, 2016
National Scholarship Ministry of Education, China, 2016