Yuhao Zhang

Hey there, welcome!

I am a founding team researcher at Samaya AI, where I currently lead the ML team. My research focuses on building AI systems that can find, reason over, and act on knowledge reliably, by improving the models, the harnesses that run them, and the environments used to evaluate them.

Before Samaya, I was a scientist at Amazon AWS AI where I worked on core AWS services relevant to enterprise search. I obtained my PhD degree from Stanford University, where I was jointly advised by Prof. Chris Manning in the Stanford NLP Group and Prof. Curtis Langlotz in the Stanford AIMI Center. My PhD work focused on foundations for natural language processing and its applications in medicine.

research interest

I care about AI systems and their impact in real-world applications.

Current focus

AI agents for knowledge-intensive work: agentic search, tool use, and long-horizon reasoning;
retrieval systems;
evaluation and benchmarks for AI agents (e.g., FrontierFinance).

Previously: information extraction, summarization, and NLP for medicine; multimodal learning; syntactic analysis and open-source NLP toolkits (I co-authored the widely used Stanza NLP library).

contact

You can reach me now at {first-name} ~at~ cs.stanford.edu. You can also find my various social accounts at the bottom of this page.

selected publications

For a complete list, see the publications page, or my google scholar page.

(*=equal contribution)

arXiv

FrontierFinance: A Challenging Benchmark for Measuring Frontier Intelligence of Finance Agents

Yuhao Zhang, O. Ozan Koyluoglu , Thejas Venkatesh , and 4 more authors

arXiv preprint arXiv:2608.11683, 2026

Bib HTML Website

@article{zhang2026frontierfinance,
  title = {FrontierFinance: A Challenging Benchmark for Measuring Frontier Intelligence of Finance Agents},
  author = {Zhang, Yuhao and Koyluoglu, O. Ozan and Venkatesh, Thejas and Diehl Martinez, Richard and Bhatia, Vishank and Alidoust, Arash and Paranjape, Ashwin},
  journal = {arXiv preprint arXiv:2608.11683},
  year = {2026},
}

COLM

Lost in the Maze: Overcoming Context Limitations in Long-Horizon Agentic Search

Howard Yen , Yoonsang Lee , Ashwin Paranjape , and 5 more authors

In Conference on Language Modeling (COLM) , 2026

Bib HTML Blog Code

@inproceedings{yen2026lost,
  title = {Lost in the Maze: Overcoming Context Limitations in Long-Horizon Agentic Search},
  author = {Yen, Howard and Lee, Yoonsang and Paranjape, Ashwin and Xia, Mengzhou and Venkatesh, Thejas and Hessel, Jack and Chen, Danqi and Zhang, Yuhao},
  booktitle = {Conference on Language Modeling (COLM)},
  year = {2026},
}

CAIS

OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction

Skyler Hallinan , Thejas Venkatesh , Xiang Ren , and 4 more authors

In Proceedings of the ACM Conference on AI and Agentic Systems (CAIS) , 2026

Bib HTML Blog

@inproceedings{hallinan2026opaquetoolsbench,
  title = {OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction},
  author = {Hallinan, Skyler and Venkatesh, Thejas and Ren, Xiang and Karimireddy, Sai Praneeth and Paranjape, Ashwin and Zhang, Yuhao and Hessel, Jack},
  booktitle = {Proceedings of the ACM Conference on AI and Agentic Systems (CAIS)},
  pages = {620--638},
  year = {2026},
  doi = {10.1145/3786335.3813182},
}

arXiv

Promptriever: Instruction-Trained Retrievers Can Be Prompted Like Language Models

Orion Weller , Benjamin Van Durme , Dawn Lawrie , and 3 more authors

arXiv preprint arXiv:2409.11136, 2024

Bib HTML

@article{weller2024promptriever,
  title = {Promptriever: Instruction-Trained Retrievers Can Be Prompted Like Language Models},
  author = {Weller, Orion and Van Durme, Benjamin and Lawrie, Dawn and Paranjape, Ashwin and Zhang, Yuhao and Hessel, Jack},
  journal = {arXiv preprint arXiv:2409.11136},
  year = {2024},
}

EMNLP

Dancing in Chains: Reconciling Instruction Following and Faithfulness in Language Models

Zhengxuan Wu , Yuhao Zhang, Peng Qi , and 6 more authors

In EMNLP , 2024

Bib HTML

@inproceedings{wu2024dancing,
  title = {Dancing in Chains: Reconciling Instruction Following and Faithfulness in Language Models},
  author = {Wu, Zhengxuan and Zhang, Yuhao and Qi, Peng and Xu, Yumo and Han, Rujun and Zhang, Yian and Chen, Jifan and Min, Bonan and Huang, Zhiheng},
  booktitle = {EMNLP},
  year = {2024},
}

EMNLP

RAG-QA Arena: Evaluating Domain Robustness for Long-form Retrieval Augmented Question Answering

Rujun Han , Yuhao Zhang, Peng Qi , and 6 more authors

In EMNLP , 2024

Bib HTML

@inproceedings{han2024rag,
  title = {RAG-QA Arena: Evaluating Domain Robustness for Long-form Retrieval Augmented Question Answering},
  author = {Han, Rujun and Zhang, Yuhao and Qi, Peng and Xu, Yumo and Wang, Jenyuan and Liu, Lan and Wang, William Yang and Min, Bonan and Castelli, Vittorio},
  booktitle = {EMNLP},
  year = {2024},
}

ACL Findings

RobustQA: Benchmarking the Robustness of Domain Adaptation for Open-domain Question Answering

Rujun Han , Peng Qi , Yuhao Zhang, and 6 more authors

In Findings of the Annual Meeting of the Association for Computational Linguistics (ACL) , 2023

Bib HTML

@inproceedings{chen2023improving,
  title = {RobustQA: Benchmarking the Robustness of Domain Adaptation for Open-domain Question Answering},
  author = {Han, Rujun and Qi, Peng and Zhang, Yuhao and Liu, Lan and Burger, Juliette and Wang, William and Huang, Zhiheng and Xiang, Bing and Roth, Dan},
  booktitle = {Findings of the Annual Meeting of the Association for Computational Linguistics (ACL)},
  year = {2023},
}

MLHC

Contrastive Learning of Medical Visual Representations from Paired Images and Text

Yuhao Zhang, Hang Jiang , Yasuhide Miura , and 2 more authors

In Proceedings of the 7th Machine Learning for Healthcare Conference , 2022

Bib HTML

@inproceedings{zhang2022contrastive,
  title = {Contrastive Learning of Medical Visual Representations from Paired Images and Text},
  author = {Zhang, Yuhao and Jiang, Hang and Miura, Yasuhide and Manning, Christopher D and Langlotz, Curtis P},
  booktitle = {Proceedings of the 7th Machine Learning for Healthcare Conference},
  pages = {1--24},
  volume = {182},
  year = {2022},
  series = {Proceedings of Machine Learning Research},
  publisher = {PMLR},
  dataset = {https://github.com/yuhaozhang/convirt},
}

Thesis

Deep Understanding and Generation of Medical Text and Beyond

Yuhao Zhang

Stanford University PhD Thesis, 2021

Bib HTML

@article{zhang2021deep,
  title = {Deep Understanding and Generation of Medical Text and Beyond},
  author = {Zhang, Yuhao},
  year = {2021},
  journal = {Stanford University PhD Thesis},
  school = {Stanford University},
}

ACL

Stanza: A Python Natural Language Processing Toolkit for Many Human Languages

Peng Qi* , Yuhao Zhang*, Yuhui Zhang , and 2 more authors

In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL): System Demonstrations , 2020

Bib HTML

@inproceedings{qi2020stanza,
  title = {Stanza: A Python Natural Language Processing Toolkit for Many Human Languages},
  author = {Qi*, Peng and Zhang*, Yuhao and Zhang, Yuhui and Bolton, Jason and Manning, Christopher D},
  booktitle = {Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL): System Demonstrations},
  year = {2020},
}