Influential Security Papers
Announcement: As of September 2025, this service will be discontinued. Google has recently introduced CAPTCHAs for standard access to Google Scholar, preventing automated crawling even within permitted data rate limits. It is sad to end the service after it has been running successfully for the past seven years, and I would like to thank all users for their support.
This webpage is an attempt to assemble a ranking of top-cited papers from the area of computer security. The ranking is automatically created based on citations of papers published at top security conferences. In particular, the ranking is based on the four tier-1 conferences (see the System Security Circus)
and the following tier-2 conferences
The citations for each paper are determined by crawling the DBLP service and Google Scholar. As both services limit crawling activity, the update interval for the ranking is large, such that citation counts change on average every two months.
Top of the Notch
Top-cited papers from 1980 to 2025 ⌄
-
1
Nicholas Carlini and David A. Wagner:
Towards Evaluating the Robustness of Neural Networks.
IEEE Symposium on Security and Privacy (S&P), 2017
7157% above average of year
Visited: Jun-2025
-
2
Martín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan, Ilya Mironov, Kunal Talwar, and Li Zhang:
Deep Learning with Differential Privacy.
ACM Conference on Computer and Communications Security (CCS), 2016
4635% above average of year
Visited: Jul-2025
-
3
Vipul Goyal, Omkant Pandey, Amit Sahai, and Brent Waters:
Attribute-based encryption for fine-grained access control of encrypted data.
ACM Conference on Computer and Communications Security (CCS), 2006
3869% above average of year
Visited: Jul-2025
-
4
Mihir Bellare and Phillip Rogaway:
Random Oracles are Practical: A Paradigm for Designing Efficient Protocols.
ACM Conference on Computer and Communications Security (CCS), 1993
4563% above average of year
Visited: Jun-2025
-
5
Roger Dingledine, Nick Mathewson, and Paul F. Syverson:
Tor: The Second-Generation Onion Router.
USENIX Security Symposium, 2004
2742% above average of year
Visited: Jul-2025
→ Check out the top-100 ranking
Absolute citations are not necessarily a good indicator for the impact of a paper, as the number of citations usually grows with the age of a paper. The following list shows an alternative ranking, where the citations are normalized by the age of each paper.
Top-cited papers normalized by age ⌄
-
1
Xinyue Shen, Zeyuan Chen, Michael Backes, Yun Shen, and Yang Zhang:
"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models.
ACM Conference on Computer and Communications Security (CCS), 2024
9008% above average of year
Visited: Jul-2025
-
2
Nicholas Carlini and David A. Wagner:
Towards Evaluating the Robustness of Neural Networks.
IEEE Symposium on Security and Privacy (S&P), 2017
7157% above average of year
Visited: Jun-2025
-
3
Gelei Deng, Yi Liu, Yuekang Li, Kailong Wang, Ying Zhang, Zefeng Li, Haoyu Wang, Tianwei Zhang, and Yang Liu:
MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots.
Network and Distributed System Security Symposium (NDSS), 2024
5073% above average of year
Visited: Jun-2025
-
4
Martín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan, Ilya Mironov, Kunal Talwar, and Li Zhang:
Deep Learning with Differential Privacy.
ACM Conference on Computer and Communications Security (CCS), 2016
4635% above average of year
Visited: Jul-2025
-
5
Mihir Bellare and Phillip Rogaway:
Random Oracles are Practical: A Paradigm for Designing Efficient Protocols.
ACM Conference on Computer and Communications Security (CCS), 1993
4563% above average of year
Visited: Jun-2025
→ Check out the normalized top-100 ranking
The Last Decades
If you are interested in a more detailed breakdown of top-cited papers over time, you can find can find rankings for the last decades here:
Limitations
As with any ranking, the presented results do not necessarily reflect the true impact of a paper. Citations are only one metric to assess the reception of a paper and are insufficient to characterize all aspects contributing to the relevance of scientific work. Moreover, the underlying data may contain errors or missing information. Errare humanum est.
Contact
If you have questions, comments, complains, or ideas how to improve this webpage, feel free to send an email to Konrad Rieck. Alternatively, you can contact me on Bluesky.