Safety in Large Reasoning Models: A Survey

Cheng Wang; Yue Liu; Baolong Bi; Duzhen Zhang; Zhong Zhi Li; Yingwei Ma; Yufei He; Shengju Yu; Xinfeng Li; Junfeng Fang; Jiaheng Zhang; Bryan Hooi

doi:10.18653/v1/2025.findings-emnlp.185

Safety in Large Reasoning Models: A Survey

Cheng Wang
, Yue Liu
, Baolong Bi
, Duzhen Zhang
, Zhong Zhi Li
, Yingwei Ma
, Yufei He
, Shengju Yu
, Xinfeng Li
, Junfeng Fang^*
, Jiaheng Zhang
, Bryan Hooi

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

1 Citation (Scopus)

Abstract

Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents the first comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies specific to these powerful reasoning-enhanced models. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.

Original language	English
Title of host publication	EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025
Editors	Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Publisher	Association for Computational Linguistics (ACL)
Pages	3468-3482
Number of pages	15
ISBN (Electronic)	9798891763357
DOIs	https://doi.org/10.18653/v1/2025.findings-emnlp.185
Publication status	Published - Nov 2025
Event	30th Conference on Empirical Methods in Natural Language Processing - Suzhou, China Duration: 4 Nov 2025 → 9 Nov 2025 https://aclanthology.org/volumes/2025.findings-emnlp/ (Conference Proceedings) https://underline.io/events/502/reception (Conference website)

Publication series

Name	EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP

Conference

Conference	30th Conference on Empirical Methods in Natural Language Processing
Abbreviated title	EMNLP 2025
Country/Territory	China
City	Suzhou
Period	4/11/25 → 9/11/25
Internet address	https://aclanthology.org/volumes/2025.findings-emnlp/ (Conference Proceedings) https://underline.io/events/502/reception (Conference website)

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.18653/v1/2025.findings-emnlp.185

Cite this

Wang, C., Liu, Y., Bi, B., Zhang, D., Li, Z. Z., Ma, Y., He, Y., Yu, S., Li, X., Fang, J., Zhang, J., & Hooi, B. (2025). Safety in Large Reasoning Models: A Survey. In C. Christodoulopoulos, T. Chakraborty, C. Rose, & V. Peng (Eds.), EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025 (pp. 3468-3482). (EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2025.findings-emnlp.185

Wang, Cheng ; Liu, Yue ; Bi, Baolong et al. / Safety in Large Reasoning Models : A Survey. EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025. editor / Christos Christodoulopoulos ; Tanmoy Chakraborty ; Carolyn Rose ; Violet Peng. Association for Computational Linguistics (ACL), 2025. pp. 3468-3482 (EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP).

@inproceedings{2e319497881444fd8b6eff0873aca045,

title = "Safety in Large Reasoning Models: A Survey",

abstract = "Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents the first comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies specific to these powerful reasoning-enhanced models. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.",

author = "Cheng Wang and Yue Liu and Baolong Bi and Duzhen Zhang and Li, \{Zhong Zhi\} and Yingwei Ma and Yufei He and Shengju Yu and Xinfeng Li and Junfeng Fang and Jiaheng Zhang and Bryan Hooi",

note = "Publisher Copyright: {\textcopyright} 2025 Association for Computational Linguistics. Funding Information: This research is supported by the Ministry of Education, Singapore, under the Academic Research Fund Tier 1 (FY2025) (Grant T1 251RES2507).; 30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025 ; Conference date: 04-11-2025 Through 09-11-2025",

year = "2025",

month = nov,

doi = "10.18653/v1/2025.findings-emnlp.185",

language = "English",

series = "EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP",

publisher = "Association for Computational Linguistics (ACL)",

pages = "3468--3482",

editor = "Christos Christodoulopoulos and Tanmoy Chakraborty and Carolyn Rose and Violet Peng",

booktitle = "EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025",

address = "Australia",

url = "https://aclanthology.org/volumes/2025.findings-emnlp/, https://underline.io/events/502/reception",

}

Wang, C, Liu, Y, Bi, B, Zhang, D, Li, ZZ, Ma, Y, He, Y, Yu, S, Li, X, Fang, J, Zhang, J & Hooi, B 2025, Safety in Large Reasoning Models: A Survey. in C Christodoulopoulos, T Chakraborty, C Rose & V Peng (eds), EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025. EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP, Association for Computational Linguistics (ACL), pp. 3468-3482, 30th Conference on Empirical Methods in Natural Language Processing, Suzhou, China, 4/11/25. https://doi.org/10.18653/v1/2025.findings-emnlp.185

Safety in Large Reasoning Models: A Survey. / Wang, Cheng; Liu, Yue; Bi, Baolong et al.
EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025. ed. / Christos Christodoulopoulos; Tanmoy Chakraborty; Carolyn Rose; Violet Peng. Association for Computational Linguistics (ACL), 2025. p. 3468-3482 (EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Safety in Large Reasoning Models

T2 - 30th Conference on Empirical Methods in Natural Language Processing

AU - Wang, Cheng

AU - Liu, Yue

AU - Bi, Baolong

AU - Zhang, Duzhen

AU - Li, Zhong Zhi

AU - Ma, Yingwei

AU - He, Yufei

AU - Yu, Shengju

AU - Li, Xinfeng

AU - Fang, Junfeng

AU - Zhang, Jiaheng

AU - Hooi, Bryan

N1 - Publisher Copyright: © 2025 Association for Computational Linguistics. Funding Information: This research is supported by the Ministry of Education, Singapore, under the Academic Research Fund Tier 1 (FY2025) (Grant T1 251RES2507).

PY - 2025/11

Y1 - 2025/11

N2 - Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents the first comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies specific to these powerful reasoning-enhanced models. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.

AB - Large Reasoning Models (LRMs) have exhibited extraordinary prowess in tasks like mathematics and coding, leveraging their advanced reasoning capabilities. Nevertheless, as these capabilities progress, significant concerns regarding their vulnerabilities and safety have arisen, which can pose challenges to their deployment and application in real-world settings. This paper presents the first comprehensive survey of LRMs, meticulously exploring and summarizing the newly emerged safety risks, attacks, and defense strategies specific to these powerful reasoning-enhanced models. By organizing these elements into a detailed taxonomy, this work aims to offer a clear and structured understanding of the current safety landscape of LRMs, facilitating future research and development to enhance the security and reliability of these powerful models.

UR - https://www.scopus.com/pages/publications/105028998146

U2 - 10.18653/v1/2025.findings-emnlp.185

DO - 10.18653/v1/2025.findings-emnlp.185

M3 - Conference proceeding

AN - SCOPUS:105028998146

T3 - EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP

SP - 3468

EP - 3482

BT - EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025

A2 - Christodoulopoulos, Christos

A2 - Chakraborty, Tanmoy

A2 - Rose, Carolyn

A2 - Peng, Violet

PB - Association for Computational Linguistics (ACL)

Y2 - 4 November 2025 through 9 November 2025

ER -

Wang C, Liu Y, Bi B, Zhang D, Li ZZ, Ma Y et al. Safety in Large Reasoning Models: A Survey. In Christodoulopoulos C, Chakraborty T, Rose C, Peng V, editors, EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025. Association for Computational Linguistics (ACL). 2025. p. 3468-3482. (EMNLP - Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP). doi: 10.18653/v1/2025.findings-emnlp.185

Safety in Large Reasoning Models: A Survey

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this