@inproceedings{zhao-etal-2026-peribench,
title = "{PERIBENCH}: A Protocol-Aware Benchmark for Dynamic Perioperative Risk Prediction from Numerical Models to Language Interfaces",
author = "Zhao, Xiaohui and
Meng, Xiangyu and
Chi, Jinjin and
Jing, Yongcheng and
Wang, Yiming and
Zhan, Yibing and
Sun, Hang and
Tao, Dacheng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1638/",
pages = "37223--37248",
ISBN = "979-8-89176-459-0",
abstract = "Dynamic perioperative risk prediction requires models to anticipate evolving physiological deterioration under strict prediction-time information boundaries. We introduce PERIBENCH, a protocol-aware benchmark built from high-frequency intraoperative signals. PERIBENCH defines 18 clinically motivated tasks spanning future risk classification, persistent deterioration prediction, continuous vital-sign regression, and case-level prognosis, and evaluates 37 systems across numerical predictors, deep sequential models, time-series foundation models, and LLM-oriented language interfaces. Rather than using a single undifferentiated leaderboard, PERIBENCH separates supervised or adapted prediction, direct pretrained transfer, and prompt-based language-interface inference into protocol-compatible pools. Experiments show that model sophistication does not translate monotonically into better perioperative prediction: tree-based references remain strong on many classification endpoints, deep sequential models are competitive on selected physiological tasks, direct foundation-model transfer is heterogeneous, and language interfaces capture useful classification signals but remain less reliable for regression, calibration, alert burden, and abnormal-region prediction. PERIBENCH provides a structured benchmark for studying dynamic perioperative prediction from numerical models to language interfaces under shared task semantics and explicit protocol boundaries. Code is available at https://github.com/Kdyyh/PERIBENCH."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2026-peribench">
<titleInfo>
<title>PERIBENCH: A Protocol-Aware Benchmark for Dynamic Perioperative Risk Prediction from Numerical Models to Language Interfaces</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiaohui</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangyu</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinjin</namePart>
<namePart type="family">Chi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yongcheng</namePart>
<namePart type="family">Jing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiming</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yibing</namePart>
<namePart type="family">Zhan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hang</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dacheng</namePart>
<namePart type="family">Tao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Dynamic perioperative risk prediction requires models to anticipate evolving physiological deterioration under strict prediction-time information boundaries. We introduce PERIBENCH, a protocol-aware benchmark built from high-frequency intraoperative signals. PERIBENCH defines 18 clinically motivated tasks spanning future risk classification, persistent deterioration prediction, continuous vital-sign regression, and case-level prognosis, and evaluates 37 systems across numerical predictors, deep sequential models, time-series foundation models, and LLM-oriented language interfaces. Rather than using a single undifferentiated leaderboard, PERIBENCH separates supervised or adapted prediction, direct pretrained transfer, and prompt-based language-interface inference into protocol-compatible pools. Experiments show that model sophistication does not translate monotonically into better perioperative prediction: tree-based references remain strong on many classification endpoints, deep sequential models are competitive on selected physiological tasks, direct foundation-model transfer is heterogeneous, and language interfaces capture useful classification signals but remain less reliable for regression, calibration, alert burden, and abnormal-region prediction. PERIBENCH provides a structured benchmark for studying dynamic perioperative prediction from numerical models to language interfaces under shared task semantics and explicit protocol boundaries. Code is available at https://github.com/Kdyyh/PERIBENCH.</abstract>
<identifier type="citekey">zhao-etal-2026-peribench</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1638/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>37223</start>
<end>37248</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PERIBENCH: A Protocol-Aware Benchmark for Dynamic Perioperative Risk Prediction from Numerical Models to Language Interfaces
%A Zhao, Xiaohui
%A Meng, Xiangyu
%A Chi, Jinjin
%A Jing, Yongcheng
%A Wang, Yiming
%A Zhan, Yibing
%A Sun, Hang
%A Tao, Dacheng
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhao-etal-2026-peribench
%X Dynamic perioperative risk prediction requires models to anticipate evolving physiological deterioration under strict prediction-time information boundaries. We introduce PERIBENCH, a protocol-aware benchmark built from high-frequency intraoperative signals. PERIBENCH defines 18 clinically motivated tasks spanning future risk classification, persistent deterioration prediction, continuous vital-sign regression, and case-level prognosis, and evaluates 37 systems across numerical predictors, deep sequential models, time-series foundation models, and LLM-oriented language interfaces. Rather than using a single undifferentiated leaderboard, PERIBENCH separates supervised or adapted prediction, direct pretrained transfer, and prompt-based language-interface inference into protocol-compatible pools. Experiments show that model sophistication does not translate monotonically into better perioperative prediction: tree-based references remain strong on many classification endpoints, deep sequential models are competitive on selected physiological tasks, direct foundation-model transfer is heterogeneous, and language interfaces capture useful classification signals but remain less reliable for regression, calibration, alert burden, and abnormal-region prediction. PERIBENCH provides a structured benchmark for studying dynamic perioperative prediction from numerical models to language interfaces under shared task semantics and explicit protocol boundaries. Code is available at https://github.com/Kdyyh/PERIBENCH.
%U https://aclanthology.org/2026.emnlp-main.1638/
%P 37223-37248
Markdown (Informal)
[PERIBENCH: A Protocol-Aware Benchmark for Dynamic Perioperative Risk Prediction from Numerical Models to Language Interfaces](https://aclanthology.org/2026.emnlp-main.1638/) (Zhao et al., EMNLP 2026)
ACL
- Xiaohui Zhao, Xiangyu Meng, Jinjin Chi, Yongcheng Jing, Yiming Wang, Yibing Zhan, Hang Sun, and Dacheng Tao. 2026. PERIBENCH: A Protocol-Aware Benchmark for Dynamic Perioperative Risk Prediction from Numerical Models to Language Interfaces. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 37223–37248, Budapest, Hungary. Association for Computational Linguistics.