Ranran Haoran Zhang

I am a PhD candidate in Computer Science and Engineering at Penn State University, advised by Dr. Rui Zhang. My research interests span two areas:

Data annotation in practice. I study how real-world annotation processes diverge from expectations, including annotator bias and guideline inconsistencies, and how to build more robust NLP models despite these gaps.
LLM inference. I work on making large language model serving faster and more efficient, with recent focus on speculative decoding.

Previously, I obtained my M.S. in Information Management from University of Illinois Urbana-Champaign, advised by Dr. Heng Ji, and my B.S. in Computer Science from Changsha University of Science & Technology, advised by Dr. Daojian Zeng.

news

Feb 14, 2026	New blog post: Rethinking vllm-metal’s Memory Budget for Apple Silicon.
Oct 01, 2025	New preprint: Batch Speculative Decoding Done Right is on arXiv.

selected publications

arXiv

Batch Speculative Decoding Done Right

Ranran Haoran Zhang, Soumik Dey, Ashirbad Mishra, and 3 more authors

arXiv preprint, 2025

arXiv Bib HTML

@article{zhang2025batchspec,
  title = {Batch Speculative Decoding Done Right},
  author = {Zhang, Ranran Haoran and Dey, Soumik and Mishra, Ashirbad and Wu, Hansi and Li, Binbin and Zhang, Rui},
  journal = {arXiv preprint},
  year = {2025},
}

NAACL

From Lazy to Prolific: Tackling Missing Labels in Open Vocabulary Extreme Classification by Positive-Unlabeled Sequence Learning

Ranran Haoran Zhang, Bensu Uçar, Soumik Dey, and 3 more authors

In Findings of the Association for Computational Linguistics: NAACL 2025, 2025

arXiv Bib HTML

@inproceedings{zhang2025lazytoprolific,
  title = {From Lazy to Prolific: Tackling Missing Labels in Open Vocabulary Extreme Classification by Positive-Unlabeled Sequence Learning},
  author = {Zhang, Ranran Haoran and U\c{c}ar, Bensu and Dey, Soumik and Wu, Hansi and Li, Binbin and Zhang, Rui},
  booktitle = {Findings of the Association for Computational Linguistics: NAACL 2025},
  year = {2025},
}

EACL

ConEntail: An Entailment-based Framework for Universal Zero and Few Shot Classification with Supervised Contrastive Pretraining

Ranran Haoran Zhang, Aysa Xuemo Fan, and Rui Zhang

In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL), 2023

Bib HTML

@inproceedings{zhang2023conentail,
  title = {ConEntail: An Entailment-based Framework for Universal Zero and Few Shot Classification with Supervised Contrastive Pretraining},
  author = {Zhang, Ranran Haoran and Fan, Aysa Xuemo and Zhang, Rui},
  booktitle = {Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL)},
  year = {2023},
}

NAACL

COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation

Qingyun Wang, Manling Li, Xuan Wang, and 22 more authors

In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL) - System Demonstrations, 2021

Best Demo Award arXiv Bib HTML

Best Demo Award

@inproceedings{wang2021covid,
  title = {COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation},
  author = {Wang, Qingyun and Li, Manling and Wang, Xuan and Parulian, Nikolaus and Han, Guangxing and Ma, Jiawei and Tu, Jingxuan and Lin, Ying and Zhang, Ranran Haoran and Liu, Weili and Chauhan, Aabhas and Guan, Yingjun and Li, Bangzheng and Li, Ruisong and Song, Xiangchen and Ji, Heng and Han, Jiawei and Chang, Shih-Fu and Pustejovsky, James and Liem, David and Elsayed, Ahmed and Palmer, Martha and Rah, Jasmine and Schneider, Cynthia and Onyshkevych, Boyan},
  booktitle = {Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL) - System Demonstrations},
  year = {2021},
}

EMNLP

Minimize Exposure Bias of Seq2Seq Models in Joint Entity and Relation Extraction

Ranran Haoran Zhang^*, Qianying Liu^*, Aysa Xuemo Fan, and 5 more authors

In Findings of the Association for Computational Linguistics: EMNLP 2020, 2020

Bib HTML

@inproceedings{zhang2020minimize,
  title = {Minimize Exposure Bias of Seq2Seq Models in Joint Entity and Relation Extraction},
  author = {Zhang, Ranran Haoran and Liu, Qianying and Fan, Aysa Xuemo and Ji, Heng and Zeng, Daojian and Cheng, Fei and Kawahara, Daisuke and Kurohashi, Sadao},
  booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2020},
  year = {2020},
}

AAAI

CopyMTL: Copy Mechanism for Joint Extraction of Entities and Relations with Multi-Task Learning

Daojian Zeng^*, Ranran Haoran Zhang^*, and Qianying Liu

In Proceedings of the AAAI Conference on Artificial Intelligence, 2020

arXiv Bib HTML

@inproceedings{zeng2020copymtl,
  title = {CopyMTL: Copy Mechanism for Joint Extraction of Entities and Relations with Multi-Task Learning},
  author = {Zeng, Daojian and Zhang, Ranran Haoran and Liu, Qianying},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2020},
}