CV

Research Assistant

The Ohio State University

Advised by Prof. Ting Zhu.

Research Assistant

University of Michigan, LIT group

Advised by Prof. Rada Mihalcea.

Research Assistant

University of Michigan

Advised by Prof. Michael P. Wellman.

Ph.D. in Computer Science and Engineering

The Ohio State University

Artificial Intelligence track

• Expected graduation Aug. 2029.

B.S. in Computer Science and Engineering

University of Michigan

B.S. in Electrical and Computer Engineering

Shanghai Jiao Tong University

Achieving Efficient Storage and Communication via Collaboration (CoOrbit)

SIGCOMM 2026 (16.2% acceptance rate)

Ruichen Li*, Yufan Wu*, Zhengyi Hu*, Sheng-Jyun Cai, Lang Wei, Qifan Yang, Ting Zhu. A storage-efficient Earth-observation pipeline that removes temporal and spatial redundancy.

CritICL: Inference-Time Weak-to-Strong Generalization from Small Language Model Failure Modes

COLM 2026

Yufan Wu, Yinghui He, Zhengyi Hu, Lang Wei, Ruichen Li, Qifan Yang, Ting Zhu. An inference-time framework that improves LLM reasoning by retrieving critiques mined from small-model failure modes, reaching higher accuracy at lower inference cost.

QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks

NeurIPS 2026

Jian Xie*, Tianhe Lin*, Zilu Wang*, …, Yufan Wu, et al. A family of 2B–35B open deep-research agents trained entirely on synthetic research tasks, spanning fact seeking and long-form report synthesis.

FLAP: Failure-Aware Planning for Search Agents

Submitted to EMNLP 2026 (ACL ARR May 2026)

Yufan Wu, Yinghui He, Zhengyi Hu, Lang Wei, Ruichen Li, Qifan Yang, Ting Zhu. A failure-aware planning framework that separates search planning from execution with evidence verification and local recovery, outperforming RL-based search agents by up to 8.5% across seven QA benchmarks.

Exploring LEO Satellite Networks for Continuous Monitoring and Dynamic Tracking (LENS)

SenSys 2026 (19.8% acceptance rate)

Lang Wei, Ruichen Li, Baodong Chen, Qifan Yang, Yufan Wu, Ting Zhu. A LEO satellite network scheduling system for continuous monitoring and event tracking.

Hi-ToM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models

EMNLP 2023 (Findings) — cited 100+

Yufan Wu*, Yinghui He*, Yilin Jia, Rada Mihalcea, Yulong Chen, Naihao Deng. A benchmark for higher-order theory-of-mind reasoning in LLMs, with an automated data generator and human annotation of 600+ samples.

Dean's List — Outstanding GPA

University of Michigan

Mathematical Contest in Modeling — Finalist Award (top 2%)

Consortium for Mathematics and its Applications

Undergraduate Merit Scholarship — Outstanding Performance (top 5%)

Shanghai Jiao Tong University