Bohan Lyu

I am an undergruduate at Tsinghua University, pursuing a dual degree in Computer Science and Economics since 2022.

In the summer of 2023, I joined Tsinghua NLP Lab and OpenBMB, advised by Prof. Zhiyuan Liu. In the summer of 2024, I start my research internship at Rose-STL-Lab, Department of CSE, UCSD, advised by Prof. Rose Yu.

Before that, I interned as a quantitative researcher at China Securities in 2023.

My research interests lie in facilitating scientific discovery with large models. My current research questions are:
1. How to enable models to learn from their interactions with the physical world and drive progress in scientific discovery?
2. How can we rigorously evaluate the capabilities of scientific agents with methods that go beyond static datasets?

I have served as a reviewer for ICLR 2025, and workshops of ICLR 2024, ICML 2024 and ACL 2024.

I am seeking PhD opportunities starting in Fall 2026. I welcome any collaboration or discussion. Please feel free to reach out!

Email / GitHub / Google Scholar / Twitter / CV

▶ Contact Info (click to see)

News

2024-12 Delivered a talk at the Challenge Cup Seminar hosted by Department of Automation, Tsinghua University.
2024-11 Presented my summer project at AAAI FSS.
2024-10 Awarded the Scholarship for Technological Innovation of 5,000 RMB.
2024-06 My team got No.1 out of 600+ teams at a Baidu Inc. Data Mining Competition.
2024-05 I became a member of 'Sparking Program', the most prestigious and selective academic organization for students at Tsinghua University.
2024-04 I led my team to win Second Place and the Newcomer Prize at Tsinghua University's Challenge Cup.
2024-04 I received a Grade A in the 'Global Exploration Initiative' and was awarded a grant of 20,000 RMB.

Research (* indicates equal contribution)

I'm interested in Natural Language Processing, AI4Science and multimodal learning.

	Adapting While Learning: Grounding LLMs for Scientific Problems with Intelligent Tool Usage Adaptation Bohan Lyu, Yadi Cao, Duncan Watson-Parris, Leon Bergen, Taylor Berg-Kirkpatrick, Rose Yu Preprint, 2024 arxiv / slides / youtube / This work proposes a fine-tuning method where LLMs internalize tool-generated solutions (World Knowledge Distillation) and learn to switch between direct answers and tool use for complex problems (Tool Usage Adaptation). It outperforms GPT-4 and Claude-3.5 across six scientific benchmarks.
	MEGA-Bench: Scaling Multimodal Evaluation to over 500 Real-World Tasks Jiacheng Chen, Tianhao Liang, Sherman Siu, Zhengqing Wang, Kai Wang, Yubo Wang, Yuansheng Ni, Wang Zhu, Ziyan Jiang, Bohan Lyu, Dongfu Jiang, Xuan He, Yuan Liu, Hexiang Hu, Xiang Yue, Wenhu Chen Under Peer Review*, 2024 arxiv / website / MEGA-Bench contains 505 multimodal tasks with diverse data sources, input/output formats, and skill requirements. The benchmark is equiped with a suite of 45 evaluation metrics to handle various output formats beyond multiple-choice questions.
	VIDEOSCORE: Building Automatic Metrics to Simulate Fine-grained Human Feedback for Video Generation Xuan He, Dongfu Jiang, Ge Zhang, Max Ku, Achint Soni, Sherman Siu, Haonan Chen, Abhraneil Chandra, Ziyan Jiang, Aaran Arulraj, Kai Wang, Quy Duc Do, Yuansheng Ni, Bohan Lyu, Yaswanth Narsupalli, Rongqi Fan, Zhiheng Lyu, Yuchen Lin, Wenhu Chen EMNLP (Main), 2024 arxiv / website / We release VIDEOFEEDBACK, the first large-scale dataset containing human-provided multi-aspect score over 37.6K synthesized videos from 11 existing video generative models. We train VIDEOSCORE based on VIDEOFEEDBACK to enable automatic video quality assessment.
	Exploring Diffusion Models’ Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks Xiaoyu Wu, Jiaru Zhang, Yang Hua, Bohan Lyu, Hao Wang, Tao Song, Haibing Guan Under Peer Review, 2024 arxiv / We apply Bayesian Neural Networks (BNNs) on Diffusion Models (DMs) with variational inference to implicitly broaden the learned distribution, and present that the learning target of the BNNs can be naturally regarded as an expectation of the diffusion loss and a further regularization with the pretrained DMs.
	Enhancing LLM’s Capabilities in Open Domains via Autonomous Tool Integration Bohan Lyu, Xin Cong, Heyang Yu, Pan Yang, Yujia Qin, Yining Ye, Yaxi Lu, Zhong Zhang, Yukun Yan, Yankai Lin, Zhiyuan Liu, Maosong Sun Under Peer Review, 2023 arxiv / Developed an autonomous agent that leverages GitHub repositories to extend its capabilities to address diverse user queries. Introduced a new agent architecture that achieved SOTA performance on SciAct.

Design and source code from Leonid Keselman's website

Bohan Lyu

News

Research (* indicates equal contribution)

Adapting While Learning: Grounding LLMs for Scientific Problems with Intelligent Tool Usage Adaptation

MEGA-Bench: Scaling Multimodal Evaluation to over 500 Real-World Tasks

VIDEOSCORE: Building Automatic Metrics to Simulate Fine-grained Human Feedback for Video Generation

Exploring Diffusion Models’ Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks

Enhancing LLM’s Capabilities in Open Domains via Autonomous Tool Integration