Zhuochun Li 「李卓淳」

---

News: 🎇

October 2026 [Exam] I passed my PhD dissertation proposal defense and became a PhD candidate (ABD)! 🎉
April 2026 [Exam] I have passed the PhD Comprehensive Exam and entered the Full-Time Dissertation Study (FTDJ) process.
March 2026 [Internship] I’ll be joining Microsoft as an Applied Scientist Intern for Summer 2026 in Redmond, WA.
January 2026 [Paper] My first-author paper "Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry" is accepted to ICLR 2026!!! Hope to see you all in Rio de Janeiro, Brazil in April. 🎊🎉
August 2025 [Paper] Our paper "Think Globally, Group Locally: Evaluating LLMs Using Multi-Lingual Word Grouping Games" is accepted to EMNLP 2025 Main Conference! 🎊
July 2025 [Internship] I’ll be joining Amazon as an Applied Scientist Intern for Fall 2025 in San Diego, CA.
May 2025 [Paper] My first-author paper "Learning from Committee: Reasoning Distillation from a Mixture of Teachers with Peer-Review" is accepted by ACL 2025!! Looking forward to presenting it at ACL 2025 on July 27–August 1st, 2025 | Vienna, Austria. I would love to connect with folks! Feel free to reach out if you're around. 🎉 🎉
April 2025 [Paper] Our paper "Reason-to-Rank: Distilling Direct and Comparative Reasoning from Large Language Models for Document Reranking" is accepted by SIGIR 2025! 🎊
March 2025 [Exam] I am happy to be informed that I passed my PhD Preliminary Exam. 🎓

---

Selected Publications

One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflows
Zhuochun Li, Youngmin Ko, Ali Keramati, Nicola Ferri, Susana Palmaz Lopez Pelaez, Liang-Chun Tsai, Yuepeng Wang, Mirco Milletari, Tuhin Kundu, Vadim Smolyakov, Kjartan Olafsson, Tommy Guy
Arxiv, Submitted to ICLR 2027

SPEAR: Distilling Domain-Adaptive Reasoning Skeletons via Sequential Symbolic Alignment in Reinforcement Learning
Zhuochun Li, Yuelyu Ji, Yiming Zeng, Daqing He
Arxiv, Submitted to NAACL 2027

Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry
Zhuochun Li, Yong Zhang, Ming Li, Yuelyu Ji, Yiming Zeng, Ning Cheng, Yun Zhu, Yanmeng Wang, Shaojun Wang, Jing Xiao, Daqing He
ICLR 2026

Think Globally, Group Locally: Evaluating LLMs Using Multi-Lingual Word Grouping Games
César Guerra-Solano, Zhuochun Li, Xiang Lorraine Li
EMNLP 2025 Main Conference

Learning from Committee: Reasoning Distillation from a Mixture of Teachers with Peer-Review
Zhuochun Li, Yuelyu Ji, Rui Meng, Daqing He
ACL 2025 Findings

Reason-to-Rank: Distilling Direct and Comparative Reasoning from Large Language Models for Document Reranking
Yuelyu Ji, Zhuochun Li, Rui Meng, Daqing He
SIGIR 2025

RAG-RLRC-LaySum at BioLaySumm: Integrating Retrieval-Augmented Generation and Readability Control for Layman Summarization of Biomedical Texts
Yuelyu Ji, Zhuochun Li, Rui Meng, Sonish Sivarajkumar, Yanshan Wang, Zeshui Yu, Hui Ji, Yushui Han, Hanyu Zeng, Daqing He
2024 Proceedings of the 23rd Workshop on Biomedical Natural Language Processing (BioNLP Workshop of ACL 2024)

SiaKey: A Method for Improving Few-shot Learning with Clinical Domain Information
Zhuochun Li, Khushboo Thaker, Daqing He
2023 IEEE-International Conference on Biomedical and Health Informatics(BHI 2023)

Domestic Violence Crisis Recognition Method based on Bi-LSTM+Attention
Zhuochun Li, Zhixiao Wang, Wenyao Yan, Min Huang, Qinyuan Fan, Xin Wang
2022 IEEE 8th Annual International Conference on Network and Information Systems for Computers (ICNISC 2022)

---

Industry Experience

[May 2026 – Jul 2026] Applied Scientist Intern, Microsoft
Manager: Tommy Guy, Dong Lei
Research on training and evaluation of multi-turn long-horizon agents in Copilot Studio.
Integrated the agent sandbox, Thinkingbox, with VERL frameworks to improve agent RL.
Open-sourced the business agent benchmark: Thinkingbox-bench, and established leaderboards on LLMs.
[Sep 2025 – Dec 2025] Applied Scientist Intern, Amazon
Mentor: Anish Acharya, Manager: Ed Vul
Research with projects of evaluation and optimization of LLM agents.
Improved natural language optimization of LLMs on the GEPA framework and Amazon Bedrock platform.
Implemented prompt learning rate and token usage controls to optimize training efficiency and cost.
[May 2025 – Aug 2025] Research Intern, Ping An Technology
Explored LLM Probing on reference-free evaluation tasks.
Extracted small LMs internal layer representations evaluation across 5 aspects on reasoning question and answer pairs.
Published ICLR 2026: "Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry".
[May 2022 – Aug 2022] Machine Learning Engineer Intern, MEDA AI
Improved the text-to-speech(TTS) model based on Tactron and trained it to 10M steps on AWS server.
Assisted constructed 3D character models and 40 facial attributes classification tasks on dataset CelebA.
[May 2021 – Aug 2021] Software Engineering Intern, Pactera Technology
Design and development of Intelligent Customer Service System.
Conducted surveys with over 100 customers, assisted designing database E-R model containing 34 tables.
Developed online semantic analysis system using Baidu voice recognition API with 80% code coverage.

---

Others

I like working out and doing sports. I regularly go to gym every week, play soccer ball and basketball at weekends. And I am a trained ping-pong player. 🏓
Also, I play the drum or go out for a ride with my favorite Subaru WRX STI when feel stressed, which can make me relax! 😁

---

Updated on Sep 2026