Yu Su (苏煜)
About Me

I'm a Distinguished Assistant Professor of Engineering at the Department of Computer Science and Engineering, The Ohio State University, where I co-direct the OSU NLP group, co-lead the Foundational AI team in the ICICLE AI Institute and lead the Machine Learning Foundations team in the Imageomics Institute. I got my PhD from University of California, Santa Barbara and my bachelor's degree from Tsinghua University, both in Computer Science. I also spent some fun time as a researcher at Microsoft Semantic Machines. I'm a 2025 Sloan Research Fellow and received several best/outstanding paper awards from CVPR and ACL.
I'm broadly interested in artificial intelligence, with a primary interest in the role of language as a vehicle for reasoning and communication. These days, I spend much of my time thinking about language agents [blog, tutorial], an emerging class of AI agents characterized by their language understanding and production capabilities.
I'm fascinated by biological intelligence and the power of natural selection, so one may find many references or direct inspirations from biological intelligence in my work. Meanwhile, biological systems also have their constraints and limitations, and I hope to develop advanced artificial intelligence to augment human intelligence. Some facets and applications of intelligence I'm currently interested in:
- Reasoning and grounding, especially in multimodal contexts [MMMU, SeeAct, Grokked Transformers, Pangu, UGround]
- Planning and world models [LLM-Planner, WebDreamer]
- Memory and non-parametric continual learning [HippoRAG]
- Benchmarking and evaluation [Mind2Web, TravelPlanner]
- AI for sciences [BioCLIP, ScienceAgentBench]
What's New
- 02/2025: Honored to receive the Sloan Research Fellowship!
- 02/2025: 3 papers accepted to CVPR'25: RoboSpatial, Finer-CAM, Prompt-CAM
- 02/2025: Guest lecture on memory, reasoning, and planning of language agents at the Berkeley Advanced LLM Agents MOOC.
- 01/2025: 4 papers accepted to ICLR'25: UGround (Oral), ScienceAgentBench, In-context Reranking, VisualAgentBench
- 12/2024: Talk at NeurIPS workshop about agent safety.
- 11/2024: Check out our tutorial "Language Agents: Foundations, Prospects, and Risks" at EMNLP'24!
- 11/2024: Talk at University of Michigan, Princeton University, LMU Munich/TU Darmstadt on language agents.
- 09/2024: 4 papers accepted to NeurIPS'24 and 1 paper to EMNLP'24: HippoRAG, Grokked Transformers, Calibrated Fine-tuning, VLM4Bio, and Middleware for LLMs.
- 09/2024: Recent services: Senior Area Chair for ICLR'25, Area Chair for ACL'24, EMNLP'24, COLM'24.
- 06/2024: Honored to receive Best Student Paper Award (BioCLIP) and Best Paper Finalist (MMMU) at CVPR'24!
- 05/2024: Thrilled to release HippoRAG and bring my passion about biological intelligence into AI!
- 05/2024: 2 papers accepted to ACL'24: tool learning through simulated trial and error and planning with LLMs (or why it's hard).
- 05/2024: New talk on a holistic and critical look at language agents at the CMU Agent Workshop.
- 05/2024: 3 papers accepted to ICML'24: SeeAct, TravelPlanner (Spotlight), MagicLens (Oral).
- 04/2024: Honored that both BioCLIP and MMMU are selected for oral presentation at CVPR'24 (90/11,532, 0.8%)!
- 02/2024: 3 papers accepted to CVPR'24: BioCLIP (Best Student Paper), MMMU (Best Paper Finalist), multimodal web agents.
- 02/2024: Excited to release TravelPlanner, a real-world planning benchmark for language agents.
- 01/2024: 5 papers accepted to ICLR'24 on knowledge conflicts in LLMs (Spotlight), MAmmoTH (Spotlight), AgentBench, Interpretable Transformer, MUFFIN.
- 01/2024: Thrilled to release SeeAct, enabling everyone to use GPT-4V-based web agents with one click.
- 12/2023: Thrilled to release BioCLIP, a vision foundation model for the tree of life.
- 11/2023: Excited to release MMMU, a new multimodal benchmark for Expert AGI.
- 10/2023: Invited talks at IJCAI, UCSD, Tsinghua, Fudan, University of Hong Kong, Pinterest, and Salesforce AI Research on language agents (slides).
- 10/2023: 3 papers accepted to EMNLP'23 on attribution evaluation of LLMs, text-to-SQL error detection, and biomedical NLP.
- 09/2023: 3 papers accepted to NeurIPS'23: Mind2Web (Spotlight), MagicBrush, and Holistic Transfer.
- 07/2023: Glad that LLM-Planner that uses LLMs for embodied agent planning got accepted to ICCV'23!
- 07/2023: Honored that our paper Pangu received the Outstanding Paper Award from ACL 2023!
- 06/2023: Invited talks at Army Research Lab and OSU RISK Institute.
- 05/2023: 6 papers accepted to ACL'23. Congrats to all the students and collaborators!
- 05/2023: Grateful to receive support from NIH R01 for our research on AI and biomedicine!
- 04/2023: Honored to receive support from President Johnson for our research on large language models!
- 03/2023: Invited talks at University of Tokyo and Amazon on grounding language models to real-world environments.
- 03/2023: Honored to receive the College of Engineering Lumley Research Award!
- 01/2023: Check out a summary of the major achievements by the OSU NLP group in 2022!
- 12/2022: Serve as Area Chair for ACL'23.
- 11/2022: Serve as Workflow Co-Chair for SIGKDD'23.
- 10/2022: Excited that our ArcaneQA paper won the Outstanding Paper Award at COLING'22 (top 15 out of 2253 submissions)!
- 10/2022: Papers on broad-coverage conversational AI and GPT-3 for biomedical information extraction accepted to EMNLP'22!
- 09/2022: New grant from ARL for knowledge-based embodied AI.
- 08/2022: Honored to receive the Distinguished Assistant Professorship of Engineering Inclusive Excellence from OSU for research and contributions towards democratizing AI!
- 08/2022: Paper on question answering over large knowledge graphs accepted to COLING'22.
- 07/2022: Serve as Senior PC member for AAAI'23.
- 07/2022: Invited talk at the DLG4NLP workshop at NAACL'22: Will Graphs Lead to the Next Breakthrough of Conversational AI?
- 06/2022: Our OSU team won the 3rd place in the inaugural Amazon Alexa Prize TaskBot Challenge! Check out our website.
- 05/2022: Thank you, Walmart and Cisco, for supporting our research!
- 04/2022: Talk at Nanjing University and JD.com on emerging frontiers of conversational AI.
- 02/2022: Paper on long-horizon vison-and-language navigation accepted to CVPR 2022.
- 02/2022: Paper on text-to-SQL generalization accepted to ACL 2022.
- 12/2021: Check out a summary of the major achievements by the OSU NLP group in 2021!
- 11/2021: Our team is selected to participate in the Alexa Prize SimBot challenge!
- 09/2021: Excited to be a part of the Imageomics Institute -- a new NSF HDR Institute dedicated for knowledge-guided machine learning for biology. I will lead the Machine Learning Foundations team.
- 08/2021: Paper on pre-trained language models with better reasoning capabilities accepted to EMNLP 2021.
- 07/2021: Excited to be a part of ICICLE -- a new NSF AI Institute dedicated to democratizing AI through AI and cyberinfrastructure innovations. I will lead the AI team with Eric Fosler-Lussier. Read more.
- 07/2021: Talk at USC/ISI and Beijing Academy of Artificial Intelligence on Emerging Frontiers of Conversational AI.
- 05/2021: Long paper on large-scale joint KB and text embedding accepted to ACL 2021.
- 03/2021: Received an Accelerator Grant from OSU TDAI on NLP for Social Media Pharmacovigilance.
- 03/2021: Short Paper on compositional generalization for neural semantic parsing accepted to NAACL-HLT 2021.
- 01/2021: Paper on non-i.i.d. generalization of question answering on knowledge bases accepted to TheWebConf 2021 (previously WWW).
- 11/2020: Will co-organize the First Workshop on Natural Language Processing for Programming at ACL-IJCNLP 2021.
- 09/2020: Will serve on the organizing committee of NAACL 2021.
- 09/2020: Super excited to share some of the work I've been working on at Microsoft Semantic Machines — Task-Oriented Dialogue as Dataflow Synthesis (TACL'20)
- 09/2020: Two long papers (learning language interfaces from use and data-to-text generation) accepted to EMNLP'20. One short paper on document classification for COVID-19 literature accepted to Findings of EMNLP.
- 05/2020: Serve as Area Chair (Conversational Bot/QA) at NLPCC'20. Serve in the Program Committee of ACL'20, KDD'20 (chair of Trustworthy Data Mining session), EMNLP'20, AAAI'21, AKBC'20, IntEx-SemPar'20.
- 04/2020: Long paper on logical natural language generation accepted to ACL 2020
- 03/2020: Thank you, Fujitsu Laboratories of America, for supporting our research!
- 01/2020: Started as Assistant Professor of Computer Science and Engineering at the Ohio State University
- 08/2019: Long paper on model-based interactive semantic parsing got accepted to EMNLP 2019
- 08/2019: Long paper on taxonomic categorization of documents got accepted to ICDM 2019
- 05/2019: Short paper on general-purpose textual relation embedding got accepted to ACL 2019
- 05/2019: Received Outstanding Dissertation Award of Computer Science from UCSB. Thank you UCSB!
- 05/2019: Check out what we are doing at Microsoft Semantic Machines (highlighted in Microsoft Build 2019)!
- 02/2019: Full paper on vocabulary selection got accepted to NAACL 2019
- 02/2019: Talk at Stanford NLP Seminar on democratizing data science with knowledge engines
- 11/2018: Full paper on zero-shot video captioning got accepted to AAAI 2019
- 10/2018: Started as researcher at Microsoft Semantic Machines in Berkeley working on conversational AI.
- 08/2018: Full paper on concept mining from text got accepted to ICDM 2018.
- 08/2018: Two long papers on dialog/semantic parsing got accepted to EMNLP 2018.
- 07/2018: Our work on natural language interfaces to APIs highlighted in Microsoft Research Blog!
- 06/2018: Serve as PC member for ACL'18, EMNLP'18, CoNLL'18, NLPCC'18, and AAAI'19.
- 04/2018: Paper "DialSQL: Dialogue Based Structured Query Generation" accepted to ACL'18 as long paper: Improve semantic parsing with dialog.
- 04/2018: Paper "Natural Language Interfaces with Fine-Grained User Interaction: A Case Study on Web APIs" accepted to SIGIR'18 as long paper.
- 03/2018: Awarded the Best Distinguished Graduate Student Lecture of UCSB CS Summit.
- 02/2018: Paper "Global Relation Embedding for Relation Extraction" accepted to NAACL-HLT'18: Robust relation extraction from text with global statistics.
- 02/2018: Talk about "Bridging the Gap between Human and Data with AI" at the University of Massachusetts, Amherst.
- 02/2018: Successfully organized the first Workshop on Knowledge Base Construction, Reasoning and Mining at Los Angeles. Check out the great invited talks and accepted papers!
- 01/2018: Talk about "Bridging the Gap between Human and Data with AI" at the Ohio State University.
- 12/2017: I will serve in the Program Committee (Research Track) of KDD'18
- 12/2017: Paper "Unsupervised Neural Categorization for Scientific Publications" accepted to SDM'18.
- 11/2017: Attended CIKM'17 in Singapore and gave a talk on natural lanugage interface and a tutorial on construction and querying of large-scale knowledge bases.
- 10/2017: Upcoming visits in China: 10.09-10.15 (Alibaba, Hangzhou), 10.10 (Fudan University, Shanghai), 10.11 (The Computing Conferencce, Hangzhou), 10.16 (Tsinghua University, Beijing), 10.17 (Toutiao AI Lab, Beijing)
- 09/2017: I'm co-organizing the First Workshop on Knowledge Base Construction, Reasoning and Mining (KBCOM'18) co-located with WSDM'18 on Feb 9, 2018 at Los Angeles. CFP is out!
- 09/2017: Finished summer internship at MSR. Flying to Copenhagen for EMNLP.
- 08/2017: I will serve in the Program Committee of WWW'18
- 08/2017: Paper on natural language interface to web API from zero user and data accepted to CIKM'17.
- 07/2017: Tutorial on Construction and Querying of Large-Scale Knowledge Bases accepted to CIKM'17. See you in Singapore!
- 06/2017: Three papers on semantic parsing/QA accepted to EMNLP'17. Thanks to my collaborators!
- 06/2017: Started summer internship in Microsoft Research
- 04/2017: I will serve in the Program Committee of CIKM'17
- 03/2017: Attended a project meeting at UIUC and gave a talk on unsupervised document categorization
- 03/2017: I will serve in the Program Committee of NLPCC'17
- 02/2017: I will serve in the Program Committee of EMNLP'17
- 01/2017: I will serve in the Program Committee of ACL'17
- 11/2016: Attended EMNLP'16 in Austin, US
- 09/2016: Our QA dataset GraphQuestions v1 is released. Check it out!
- 09/2016: Two papers on knowledge base question answering got accepted to EMNLP'16!
- 09/2016: Attended the Bay Area Deep Learning School, Stanford
- 06/2016: Started summer internship in Microsoft Research, Redmond
Recent Talks
-
Tutorial: Language Agents: Foundations, Prospects, and Risks (slides) (recording)
EMNLP 2024 (w/ Diyi Yang, Shunyu Yao, Tao Yu) -
On Memory, Reasoning, and Planning of Language Agents (recording)
Berkeley Advanced LLM Agents MOOC -
A Holistic and Critical Look at Language Agents (slides)
CMU Agent Workshop, JPMorgan Chase, Amazon AGI, Apple NLU Workshop, LMU Munich/TU Darmstadt, Stanford, Princeton, NVIDIA -
Web Agents: A New Frontier for Embodied Agents (slides)
University of Michigan, SpLU-RoboNLP Workshop@ACL'24, ServiceNow
Awards & Honors
- Alfred P. Sloan Research Fellowship, 2025
- Lumley Interdisciplinary Research Award, OSU, 2025
- Best Student Paper Award, CVPR, 2024
- Best Paper Finalist, CVPR, 2024
- Cisco Faculty Award, 2024
- Outstanding Area Chair, EMNLP, 2024
- Outstanding Paper Award, ACL, 2023
- Lumley Research Award, OSU, 2023
- Outstanding Paper Award, COLING, 2022
- Distinguished Assistant Professorship of Engineering Inclusive Excellence, OSU, 2022
- Third-Place Honor, Inaugural Amazon Alexa Prize TaskBot Challenge, 2022
- Outstanding Dissertation Award of Computer Science, UCSB, 2019
- Outstanding Freshman/Graduate Awards, Tsinghua University, 2008/2012
Students
Current Ph.D. / Postdocs
- Yu Gu (SP20 –)
- Bernal Jiménez Gutiérrez (SP20 –)
- Vardaan Pahuja (SP20 –)
- Chan Hee (Luke) Song (AU20 –)
- Shijie Chen (AU21 –, co-advised with Huan Sun)
- Sam Stevens (AU21 –)
- Kai Zhang (AU22 –)
- Jiaman (Lisa) Wu (AU22 –, co-advised with Wei-Lun Chao)
- Boyuan Zheng (AU23 –)
- Yiheng Shu (AU23 –)
- Boyu Gou (AU23 –)
- Jianyang Gu (AU24 –, postdoc with Wei-lun Chao, Tanya Berger-Wolf)
Publications
-
From RAG to Memory: Non-Parametric Continual Learning for Large Language Models
Bernal Jiménez Gutiérrez, Yiheng Shu, Weijian Qi, Sizhe Zhou, Yu Su. [preprint] -
Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents
Yu Gu*, Kai Zhang*, Yuting Ning*, Boyuan Zheng*, Boyu Gou, Tianci Xue, Cheng Chang, Sanjari Srivastava, Yanan Xie, Peng Qi, Huan Sun, Yu Su. (*: Equal Contribution) [preprint] -
RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics
Chan Hee Song, Valts Blukis, Jonathan Tremblay, Stephen Tyree, Yu Su, Stan Birchfield. In the Conference on Computer Vision and Pattern Recognition, 2025 (CVPR'25) [paper] -
Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents
Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, Yu Su. In the International Conference on Learning Representations, 2025 (ICLR'25 Oral) [website] [paper] [code] -
HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models
Bernal Jiménez Gutiérrez, Yiheng Shu, Yu Gu, Michihiro Yasunaga, Yu Su. In the Conference on Neural Information Processing Systems, 2024 (NeurIPS'24) [paper] [code] -
Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization
Boshi Wang, Xiang Yue, Yu Su, Huan Sun. In the Conference on Neural Information Processing Systems, 2024 (NeurIPS'24) [paper] [code] -
MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions
Kai Zhang, Yi Luan, Hexiang Hu, Kenton Lee, Siyuan Qiao, Wenhu Chen, Yu Su, Ming-Wei Chang. In the International Conference on Machine Learning, 2024 (ICML'24 Oral) [website][paper] -
GPT-4V(ision) is a Generalist Web Agent, if Grounded
Boyuan Zheng, Boyu Gou, Jihyung Kil, Huan Sun, Yu Su. In the International Conference on Machine Learning, 2024 (ICML'24) [website] [paper] [code] -
TravelPlanner: A Benchmark for Real-World Planning with Language Agents
Jian Xie*, Kai Zhang*, Jiangjie Chen, Tinghui Zhu, Renze Lou, Yuandong Tian, Yanghua Xiao, Yu Su. In the International Conference on Machine Learning, 2024 (ICML'24 Spotlight) [website] [paper] [code] (*: Equal Contribution) -
BioCLIP: A Vision Foundation Model for the Tree of Life
Samuel Stevens*, Jiaman Wu*, Matthew J Thompson, Elizabeth G Campolongo, Chan Hee Song, David Edward Carlyn, Li Dong, Wasila M Dahdul, Charles Stewart, Tanya Berger-Wolf, Wei-Lun Chao, Yu Su. In the Conference on Computer Vision and Pattern Recognition, 2024 (CVPR'24) [website] [paper] [code] (*: Equal Contribution)
Best Student Paper -
MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
Xiang Yue*, Yuansheng Ni, Kai Zhang, Tianyu Zheng, Ruoqi Liu, Ge Zhang, Samuel Stevens, Dongfu Jiang, Weiming Ren, Yuxuan Sun, Cong Wei, Botao Yu, Ruibin Yuan, Renliang Sun, Ming Yin, Boyuan Zheng, Zhenzhu Yang, Yibo Liu, Wenhao Huang, Huan Sun, Yu Su*, Wenhu Chen*. In the Conference on Computer Vision and Pattern Recognition, 2024 (CVPR'24) [website] [paper] [code] [data] (*: corresponding authors)
Best Paper Finalist -
Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts
Jian Xie*, Kai Zhang*, Jiangjie Chen, Renze Lou, Yu Su. In the International Conference on Learning Representations, 2024 (ICLR'24 Spotlight) [paper] [code] (*: Equal Contribution) -
Mind2Web: Towards a Generalist Agent for the Web
Xiang Deng, Yu Gu, Boyuan Zheng, Shijie Chen, Samuel Stevens, Boshi Wang, Huan Sun, Yu Su. In the Conference on Neural Information Processing Systems, Datasets and Benchmarks Track, 2023 (NeurIPS'23 Spotlight) [paper] [website] [code] -
LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models
Chan Hee Song, Jiaman Wu, Clayton Washington, Brian M. Sadler, Wei-Lun Chao, Yu Su. In the International Conference on Computer Vision, 2023 (ICCV'23) [paper] [website] [code] -
Don't Generate, Discriminate: A Proposal for Grounding Language Models to Real-World Environments
Yu Gu, Xiang Deng, Yu Su. In the Annual Conference of the Association for Computational Linguistics, 2023 (ACL'23) [paper] [code] Outstanding Paper Award
Teaching
- AU2022-2024: CSE6521 - Introduction to Artificial Intelligence (Graduate)
- SP2022: CSE5525 - Foundations of Speech and Language Processing (Undergrad & Graduate)
- AU2021: CSE6521 - Introduction to Artificial Intelligence (Graduate)
- AU2020: CSE5539 - Cutting-Edge Topics in Natural Language Processing (Undergrad & Graduate)
- SP2020: CSE5243 - Introduction to Data Mining (Undergrad & Graduate)