Yushi Bai

Yushi Bai 白雨石

I am a Fourth-year Ph.D. student at Tsinghua University, advised by Prof. Juanzi Li. I reveived my B.E. in Yao class from Tsinghua University in 2022. I have been very fortunate to work with Prof. Jie Tang, Prof. Zhiyuan Liu, Prof. Jure Leskovec, and Prof. Ariel Procaccia.

Email / Google Scholar / Github / News / Selected Publications / Experience / Honors & Awards

News

[2025.09]Check out our newest paper SIRI: Scaling Iterative Reinforcement Learning with Interleaved Compression, a simple yet powerful recipe for training reasoning models that achieves both higher accuracy and fewer tokens.

[2025.09]Two papers accepted to NeurIPS 2025, including one Spotlight!

[2025.08]GLM-4.5 Technical Report is released! I am one of the core contributors for the post-training stage.

[2025.05]Three papers including LongBench v2 accepted to ACL 2025! See you in Vienna!

Selected Publications

My research interests primarily lie in large language models (LLMs), covering long-context LLM, LLM evaluation and safety, and multimodality. Recently, I have been focusing on designing more efficient and comprehensive LLM architectures that push beyond traditional architectures, particularly in long-context and agentic scenarios. Below are some of my lead works (see my Google Scholar for a complete list).

Long-Context LLM

	LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning Yuhao Wu, Yushi Bai, Zhiqiang Hu, Roy Ka-Wei Lee, Juanzi Li arxiv [paper] [code]
	LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks Yushi Bai, Shangqing Tu, Jiajie Zhang, Hao Peng, Xiaozhi Wang, Xin Lv, Shulin Cao, Jiazheng Xu, Lei Hou, Yuxiao Dong, Jie Tang, Juanzi Li ACL 2025 [paper] [website] [code] Top-15 Influential paper at ACL 25
	LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs Yushi Bai, Jiajie Zhang, Xin Lv, Linzhi Zheng, Siqi Zhu, Lei Hou, Yuxiao Dong, Jie Tang, Juanzi Li ICLR 2025 [paper] [code]
	LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA Jiajie Zhang, Yushi Bai, Xin Lv, Wanjun Gu, Danqing Liu, Minhao Zou, Shulin Cao, Lei Hou, Yuxiao Dong, Ling Feng, Juanzi Li EMNLP 2025 [paper] [code]
	LongAlign: A Recipe for Long Context Alignment of Large Language Models Yushi Bai, Xin Lv, Jiajie Zhang, Yuze He, Ji Qi, Lei Hou, Jie Tang, Yuxiao Dong, Juanzi Li EMNLP 2024 [paper] [code]
	LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding Yushi Bai, Xin Lv, Jiajie Zhang, Hongchang Lyu, Jiankai Tang, Zhidian Huang, Zhengxiao Du, Xiao Liu, Aohan Zeng, Lei Hou, Yuxiao Dong, Jie Tang, Juanzi Li ACL 2024 [paper] [code] Top-2 Influential paper at ACL 24

LLM Evaluation & Safety

	Automating dataset updates towards reliable and timely evaluation of large language models Jiahao Ying, Yixin Cao, Yushi Bai, Qianru Sun, Bo Wang, Wei Tang, Zhaojun Ding, Yizhe Yang, Xuanjing Huang, Shuicheng Yan NeurIPS 2024 [paper] [website]
	PrivacyMind: Large Language Models Can Be Contextual Privacy Protection Learners Yijia Xiao, Yiqiao Jin, Yushi Bai, Yue Wu, Xianjun Yang, Xiao Luo, Wenchao Yu, Xujiang Zhao, Yanchi Liu, Quanquan Gu, Haifeng Chen, Wei Wang, Wei Cheng EMNLP 2024 [paper] [code]
	WaterBench: Towards Holistic Evaluation of Watermarks for Large Language Models Shangqing Tu, Yuliang Sun, Yushi Bai, Jifan Yu, Lei Hou, Juanzi Li ACL 2024 [paper] [code]
	KoLA: Carefully Benchmarking World Knowledge of Large Language Models Jifan Yu, Xiaozhi Wang, Shangqing Tu, ... Yushi Bai*, ... Bin Xu, Jie Tang, Juanzi Li ICLR 2024 [paper] [website]
	Benchmarking Foundation Models with Language-Model-as-an-Examiner Yushi Bai, Jiahao Ying, Yixin Cao, Xin Lv, Yuze He, Xiaozhi Wang, Jifan Yu, Kaisheng Zeng, Yijia Xiao, Haozhe Lyu, Jiayin Zhang, Juanzi Li, Lei Hou NeurIPS 2023 [paper] [website]

Multimodal

	MMGeoLM: Hard Negative Contrastive Learning for Fine-Grained Geometric Understanding in Large Multimodal Models Kai Sun, Yushi Bai, Zhen Yang, Jiajie Zhang, Ji Qi, Lei Hou, Juanzi Li arxiv [paper] [code]
	Quicksviewer: An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes Ji Qi, Yuan Yao, Yushi Bai, Bin Xu, Juanzi Li, Zhiyuan Liu, Tat-Seng Chua arxiv [paper] [website]
	T^3Bench: Benchmarking Current Progress in Text-to-3D Generation Yuze He, Yushi Bai, Matthieu Lin, Wang Zhao, Yubin Hu, Jenny Sheng, Ran Yi, Juanzi Li, Yong-Jin Liu arxiv [paper] [website]
	CogCoM: A Visual Language Model with Chain-of-Manipulations Reasoning Ji Qi, Ming Ding, Weihan Wang, Yushi Bai, Qingsong Lv, Wenyi Hong, Bin Xu, Lei Hou, Juanzi Li, Yuxiao Dong, Jie Tang ICLR 2025 [paper] [code]
	MM-MATH: Advancing Multimodal Math Evaluation with Process Evaluation and Fine-grained Classification Kai Sun, Yushi Bai, Ji Qi, Lei Hou, Juanzi Li EMNLP 2024 [paper] [code]

Knowledge Graph Reasoning

	Answering Complex Logical Queries on Knowledge Graphs via Query Computation Tree Optimization Yushi Bai, Xin Lv, Juanzi Li, Lei Hou ICML 2023 [paper] [code]
	SQUIRE: A Sequence-to-sequence Framework for Multi-hop Knowledge Graph Reasoning Yushi Bai, Xin Lv, Juanzi Li, Lei Hou, Yincen Qu, Zelin Dai, Feiyu Xiong EMNLP 2022 (Oral) [paper] [code]
	Modeling heterogeneous hierarchies with relation-specific hyperbolic cones Yushi Bai, Rex Ying, Hongyu Ren, Jure Leskovec NeurIPS 2021 [paper] [code]

Fair Division

	Envy-free and pareto-optimal allocations for agents with asymmetric random valuations Yushi Bai, Paul Gölz IJCAI 2022 [paper]
	Fair allocations for smoothed utilities Yushi Bai, Uriel Feige, Paul Gölz, Ariel Procaccia ACM EC 2022 [paper]

Experience

	Tsinghua University 2022.08 - Present Ph.D student in Computer Science and Technology Advisor: Prof. Juanzi Li
	Harvard University 2021.02-2021.09 Visiting Research Student Advisor: Prof. Ariel Procaccia
	Stanford University 2020.06-2020.10 Visiting Research Student Advisor: Prof. Jure Leskovec
	Tsinghua University 2018.08 - 2022.07 B.Eng. in Yao Class

Honors & Awards

National Scholarship, 2024
Outstanding Undergraduate in Beijing, 2022
Yao Award, 2021
Tsinghua Comprehensive Excellence Scholarship, 2019, 2020, 2021, 2023

Template adapted from Jon Barron.