Yifan Zhang

Ph.D. student at Princeton University, focusing on Large Language Models, especially Language Modeling and Pretraining, LLM Reasoning and Reinforcement Learning

About Me

I am a Ph.D. student at Princeton University, where my research focuses on building scalable and capable large language models (LLMs). My work explores how to improve LLM reasoning, align their behavior with human preferences through general preference models, and develop new attention mechanisms and model architectures.

Previously, I was fortunate to study and conduct research at the Institute for Interdisciplinary Information Sciences (IIIS) at Tsinghua University and at the UCLA AGI Lab.

Yifan Zhang

Research Interests

  • Machine Learning and Large Language Models
  • Language Modeling and Pretraining
  • LLM Reasoning and Reinforcement Learning

You can find my publications on Google Scholar.

Selected Works

On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning

Yifan Zhang*, Yifeng Liu*, Huizhuo Yuan, Yang Yuan, Quanquan Gu, Andrew C Yao

arXiv:2505.17508

Tensor Product Attention Is All You Need

Yifan Zhang*, Yifeng Liu*, Huizhuo Yuan, Zhen Qin, Yang Yuan, Quanquan Gu, Andrew C Yao

ICML 2025 ES-FoMo Workshop; arXiv:2501.06425

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

Yifan Zhang*, Ge Zhang*, Yue Wu*, Kangping Xu, Quanquan Gu

ICML 2025

Autonomous Data Selection with Zero-shot Generative Classifiers for Mathematical Texts

Yifan Zhang*, Yifan Luo*, Yang Yuan, Andrew C Yao

ACL 2025 Findings

Augmenting Math Word Problems via Iterative Question Composing

Haoxiong Liu*, Yifan Zhang*, Yifan Luo, Andrew C Yao

AAAI 2025

Cumulative Reasoning with Large Language Models

Yifan Zhang, Jingqin Yang, Yang Yuan, Andrew C Yao

Transactions on Machine Learning Research (TMLR)

(* denotes equal contribution)

Publications

Autonomous Data Selection with Zero-shot Generative Classifiers for Mathematical Texts

Yifan Zhang*, Yifan Luo*, Yang Yuan, Andrew C Yao

ACL 2025 Findings

Tensor Product Attention Is All You Need

Yifan Zhang*, Yifeng Liu*, Huizhuo Yuan, Zhen Qin, Yang Yuan, Quanquan Gu, Andrew C Yao

ICML 2025 ES-FoMo Workshop; arXiv:2501.06425

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

Yifan Zhang*, Ge Zhang*, Yue Wu*, Kangping Xu, Quanquan Gu

ICML 2025

Augmenting Math Word Problems via Iterative Question Composing

Haoxiong Liu*, Yifan Zhang*, Yifan Luo, Andrew C Yao

AAAI 2025

Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networks

Rui Hu*, Yifan Zhang*, Zhuoran Li, Longbo Huang

ICLR 2025 Spotlight

Training and Evaluating Language Models with Template-based Data Generation

Yifan Zhang

ICLR 2025 DATA-FM Workshop; arXiv:2411.18104

SEAL: Simultaneous Label Hierarchy Exploration and Learning

Zhiquan Tan*, Zihao Wang*, Yifan Zhang*

Transactions on Machine Learning Research (TMLR)

Information Flow in Self-Supervised Learning

Zhiquan Tan, Jingqin Yang, Weiran Huang, Yang Yuan†, Yifan Zhang†

ICML 2024

Matrix Information Theory for Self-Supervised Learning

Yifan Zhang*, Jingqin Yang*, Zhiquan Tan*, Weiran Huang, Yang Yuan

ICML 2024

Meta prompting for ai systems

Yifan Zhang, Yang Yuan, Andrew C Yao

ICLR 2024 BGPT Workshop; arXiv:2311.11482

Cumulative Reasoning with Large Language Models

Yifan Zhang*, Jingqin Yang*, Yang Yuan, Andrew C Yao

Transactions on Machine Learning Research (TMLR)

EffCause: Discover Dynamic Causal Relationships Efficiently from Time-Series

Yicheng Pan, Yifan Zhang, Xinrui Jiang, Meng Ma, Ping Wang

TKDD 2024

Contrastive Learning Is Spectral Clustering On Similarity Graph

Zhiquan Tan*, Yifan Zhang*, Jingqin Yang*, Yang Yuan

ICLR 2024

Coded real number matrix multiplication for on-device edge computing

Zhiquan Tan, Dingli Yuan, Yifan Zhang, Zhongyi Huang

SPL 2023

Trade-off Between Efficiency and Consistency for Removal-based Explanations

Yifan Zhang, Haowei He*, Zhiquan Tan, Yang Yuan

NeurIPS 2023

(* denotes equal contribution, † denotes corresponding authors)

Preprints & Technical Reports

CriticLean: Critic-Guided Reinforcement Learning for Mathematical Formalization

Zhongyuan Peng, Yifan Yao, Kaijing Ma, Shuyue Guo, Yizhe Li, Yichi Zhang, Chenchen Zhang, Yifan Zhang, Zhouliang Yu, Luming Li, Minghao Liu, Yihang Xia, Jiawei Shen, Yuchen Wu, Yixin Cao, Zhaoxiang Zhang, Wenhao Huang, Jiaheng Liu, Ge Zhang

arXiv:2507.06181

On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning

Yifan Zhang*, Yifeng Liu*, Huizhuo Yuan, Yang Yuan, Quanquan Gu, Andrew C Yao

arXiv:2505.17508

FormalMATH: Benchmarking Formal Mathematical Reasoning of Large Language Models

Zhouliang Yu*, Ruotian Peng*, Keyi Ding*, Yizhe Li, Zhongyuan Peng, Minghao Liu, Yifan Zhang, Zheng Yuan, Huajian Xin, Wenhao Huang, Yandong Wen, Ge Zhang, Weiyang Liu

arXiv:2505.02735

Scaling Image Tokenizers with Grouped Spherical Quantization

Jiangtao Wang, Zhen Qin, Yifan Zhang, Tao Hu, Björn Ommer, Rania Briq, Stefan Kesselheim

arXiv:2412.02632

On the Diagram of Thought

Yifan Zhang, Yang Yuan, Andrew C Yao

arXiv:2409.10038

StackMathQA: A Curated Collection of 2 Million Mathematical Questions and Answers Sourced from Stack Exchange

Yifan Zhang

Technical Report

(* denotes equal contribution, † denotes corresponding authors)

Professional Activities

Teaching

  • Teaching Assistant, Machine Learning, IIIS, Tsinghua University

Academic Services

  • Conference Reviewer: ICLR, ICML, NeurIPS, AAAI, AISTATS
  • Journal Reviewer: TKDD, Neural Networks, Neural Computing

Invited Talks

  • Invited talk on General Preference Modeling with Preference Embeddings, November 2024
  • Invited talk on Cumulative Reasoning with Large Language Models, September 2023
  • Invited talk on Contrastive Learning Is Spectral Clustering, May 2023