Hanbing Liu

I'm a final year Thesis-track Master's student at Gaoling School of Artificial Intelligence, Renmin University of China (RUC), adviced by Prof. Qi Qi. My recent research focuses on intelligent decision-making and mechanism design in generative AI, particularly when foundation models are involved.

Prior to my graduate study, I obtined BE in Computer Science and BA in Finance (Experimental dual degree program) from the same university, where I was fortunate to work with Prof. Jing Zhang and Prof. Ju Fan on various topics around large language models.

Email / Scholar / LinkedIn

Research

Featured Preprints

Token-Level Advertising
Hanbing Liu, Bowei Zhang, Changyuan Yu, Yinyu Ye, Qi Qi†
Incentive-Based AI Alignment Workshop @ EC, 2026
arXiv / Poster / Slides (EN/CN)

Where do commercial opportunities arise in the age of generative search?

We propose a token-level auction that integrates advertiser influence into language generation through a latent mixture of advertiser policies.

When Everyone Can Submit: Designing Contests with Transparent Pre-Selection: paper figure

When Everyone Can Submit:
Designing Contests with Transparent Pre-Selection
Hanbing Liu, Ningyuan Li, Weian Li, Qi Qi†, Changyuan Yu
NeurIPS, 2026
arXiv

How should crowdsourcing platforms respond to a surge in AI-generated submissions?

We study contests with transparent pre-selection, characterizing equilibrium effort and optimal prize allocation when a cut-off score is disclosed.

Towards Budget-Aware Agentic AI: Intention-Based Planning for Costly Tool Use: paper figure

Towards Budget-Aware Agentic AI:
Intention-Based Planning for Costly Tool Use
Hanbing Liu, Chunhao Tian, Nan An, Ziyuan Wang, Pinyan Lu, Changyuan Yu, Qi Qi†
Poster @ EC, 2026
arXiv

Can we trust autonomous agents when money is at stake?

We develop intention-based planning for tool-using agents to improve task success while respecting a hard monetary budget.

Selected Publications

Less is More: Optimal Contest Design with a Shortlist
Hanbing Liu, Ningyuan Li, Weian Li, Qi Qi†, Changyuan Yu
WINE, 2025
arXiv / Poster / Slides

The optimal contest design for total effort objective is always to admit kn contestants (k ≤ 31.62%), and to award prize equally among the top kn-1 contestants.

LeVer smart contract synthesis and verification framework

Towards Trustworthy Smart Contract Synthesis:
A Multi-Agent Framework with Lean-Based Verification
Bowei Zhang*, Hanbing Liu*, Qixin Tian, Siyu Chen, Ziyuan Wang, Qi Qi†
ACL, 2026
ACL Anthology / Github

A multi-agent framework combining smart contract synthesis with Lean-based verification and adversarial testing in an iterative generate, verify, attack, and repair loop.

Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL
Hanbing Liu*, Haoyang Li*, Xiaokang Zhang, Ruotong Chen, Haiyong Xu, Tian Tian, Qi Qi, Jing Zhang†
ACL, 2025
arXiv / Poster / Github

Introducing direct preference optimization (DPO) into Text-to-SQL training pipeline, and identifing the crutial role of chain-of-thought in DPO for complex reasoning tasks.

CodeS: Towards Building Open-source Language Models for Text-to-SQL
Haoyang Li, Jing Zhang†, Hanbing Liu, Ju Fan, Xiaokang Zhang, Jun Zhu, Renjie Wei, Hongyan Pan, Cuiping Li, Hong Chen
SIGMOD, 2024
🏆 Most Influential Papers in SIGMOD'24 (1/213) [Paper Digest]
arXiv / Github / Hugging Face

The first open-source Text-to-SQL large language model and automatic domain adaptation pipeline, achieving SOTA performance.

ChatPipe: Orchestrating Data Preparation Program by Optimizing Human-ChatGPT Interactions
Sibei Chen, Hanbing Liu, Weiting Jin, Xiangyu Sun, Xiaoyao Feng, Ju Fan†, Xiaoyong Du, Nan Tang
SIGMOD, 2024 (Demo)
arXiv / Youtube

A semi-automatic machine learning programming workflow powered by reinforcement learning and ChatGPT.

α-β: Alphabetical Order, *: Equal Contribution, †: Corresponding Author

Selected Awards

Graduate Academic Excellence Scholarship, First Class, 2025
Graduate Entrance Scholarship, First Class (1st Place), 2024
Outstanding Graduate of Beijing, 2024
Outstanding Graduate of Renmin University of China, 2024
Industrial and Commercial Bank of China (ICBC) Scholarship, 2024
Huawei Scholarship, 2022,2023
Academic Excellence Scholarship, First Class (Top 3% GPA), 2022
Renmin University of China Programming Contest, Gold Medal (2nd Place), 2020
National Olympiad in Informatics in Provinces (NOIP), First Prize, 2018

Services

PC Member ICLR'27, AAAI'27, NeurIPS'26, EC'26, ICML'26 (Gold Reviewer), TMLR
Conference (Sub-)Reviewer KDD'24-27, WINE'24-26, ARR'26, AAMAS'26, ICLR'26, COCOON'25, IJTCS'25, EC'25
Workshop Reviewer EconML @ NeurIPS'26, LASS @ CIKM'26, AIMS @ ICLR'26

Great thanks to Jon Barron for the original design and implementation of this website.
Last update: Sep 21, 2026