{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T11:37:45Z","timestamp":1785497865090,"version":"3.56.0"},"reference-count":100,"publisher":"Association for Computing Machinery (ACM)","issue":"FSE","license":[{"start":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T00:00:00Z","timestamp":1782777600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"Institute of Information & Communications Technology Planning & Evaluation","award":["RS-2024-00337414"],"award-info":[{"award-number":["RS-2024-00337414"]}]},{"name":"Institute of Information & Communications Technology Planning & Evaluation","award":["RS-2024-00437306"],"award-info":[{"award-number":["RS-2024-00437306"]}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"crossref","award":["RS-2025-02293072"],"award-info":[{"award-number":["RS-2025-02293072"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Softw. Eng."],"published-print":{"date-parts":[[2026,6,30]]},"abstract":"<jats:p>Binary code analysis plays an essential role in cybersecurity, facilitating reverse engineering to reveal the inner workings of programs in the absence of source code. Traditional approaches, such as static and dynamic analysis, extract valuable insights from stripped binaries, but often demand substantial expertise and manual effort. Recent advances in deep learning have opened promising opportunities to enhance binary analysis by capturing latent features and disclosing underlying code semantics. Despite the growing number of binary analysis models based on machine learning, their robustness to adversarial code transformations at the binary level remains underexplored to date. In this work, we evaluate the robustness of deep learning models for the task of binary code similarity detection (BCSD) under semantics-preserving transformations. The unique nature of machine instructions presents distinct challenges compared to the typical input perturbations found in other domains. To achieve our goal, we introduce asmFooler, a system that evaluates the resilience of BCSD models using a diverse set of adversarial code transformations that preserve functional semantics. We construct a dataset of 9,565 binary variants from 620 baseline samples by applying eight semantics-preserving transformations across six representative BCSD models. Our major findings highlight several key insights: i) model robustness highly relies on the design of the processing pipeline, including code pre-processing, model architecture, and internal feature selection, which collectively determine how code semantics are captured; ii) the effectiveness of adversarial transformations is bounded by a transformation budget, shaped by model-specific constraints such as input size limits and the expressive capacity of semantically equivalent instructions; iii) well-crafted adversarial transformations can be highly effective, even when introducing minimal perturbations; and iv) such transformations efficiently disrupt the model\u2019s decision (e.g., misleading to false positives or false negatives) by focusing on semantically significant instructions.<\/jats:p>","DOI":"10.1145\/3797116","type":"journal-article","created":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:06:14Z","timestamp":1782839174000},"page":"1957-1980","source":"Crossref","is-referenced-by-count":1,"title":["Fool Me If You Can: On the Robustness of Binary Code Similarity Detection Models against Semantics-Preserving Transformations"],"prefix":"10.1145","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-4955-662X","authenticated-orcid":false,"given":"Jiyong","family":"Uhm","sequence":"first","affiliation":[{"name":"Sungkyunkwan University, Suwon, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-8993-3984","authenticated-orcid":false,"given":"Minseok","family":"Kim","sequence":"additional","affiliation":[{"name":"Sungkyunkwan University, Suwon, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3106-0343","authenticated-orcid":false,"given":"Michalis","family":"Polychronakis","sequence":"additional","affiliation":[{"name":"Stony Brook University, Stony Brook, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0799-0230","authenticated-orcid":false,"given":"Hyungjoon","family":"Koo","sequence":"additional","affiliation":[{"name":"Sungkyunkwan University, Suwon, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,6,30]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1145\/3243734.3243738"},{"key":"e_1_2_1_2_1","unstructured":"National Security Agency. 2025. Ghidra. https:\/\/ghidra-sre.org\/"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/3564625.3567975"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/1966913.1966919"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1007\/11790754_8"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/1315245.1315286"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.5555\/2831143.2831160"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3488204"},{"key":"e_1_2_1_9_1","unstructured":"Capstone. 2025. Capstone The Ultimate Disassembler. https:\/\/www.capstone-engine.org\/"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2017.49"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/TDSC.2013.40"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/2950290.2950350"},{"key":"e_1_2_1_13_1","volume-title":"Proceedings of the 31st USENIX Security Symposium (Security '22)","author":"Chen Qibin","year":"2022","unstructured":"Qibin Chen, Jeremy Lacomis, Edward J Schwartz, Claire Le Goues, Graham Neubig, and Bogdan Vasilescu. 2022. Augmenting decompiler output with learned variable names and types. In Proceedings of the 31st USENIX Security Symposium (Security '22)."},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.5555\/3620237.3620301"},{"key":"e_1_2_1_15_1","volume-title":"SPEC CPU 2006","author":"Standard Performance Evaluation Corporation","year":"2025","unstructured":"Standard Performance Evaluation Corporation. 2025. SPEC CPU 2006. https:\/\/www.spec.org\/cpu2006\/"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2019.00003"},{"key":"e_1_2_1_17_1","volume-title":"Proceedings of the 30th USENIX Security Symposium (Security '21)","author":"Downing Evan","year":"2021","unstructured":"Evan Downing, Yisroel Mirsky, Kyuhong Park, and Wenke Lee. 2021. DeepReflect: Discovering malicious functionality through binary reconstruction. In Proceedings of the 30th USENIX Security Symposium (Security '21)."},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3576915.3623098"},{"key":"e_1_2_1_19_1","volume-title":"Proceedings of the 23rd USENIX Security Symposium (Security '14)","author":"Egele Manuel","year":"2014","unstructured":"Manuel Egele, Maverick Woo, Peter Chapman, and David Brumley. 2014. Blanket execution: Dynamic similarity testing for program binaries and components. In Proceedings of the 23rd USENIX Security Symposium (Security '14)."},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2016.23185"},{"key":"e_1_2_1_21_1","unstructured":"Chris Evans. 2025. vsftpd. https:\/\/security.appspot.com\/vsftpd.html"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3052973.3052995"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978370"},{"key":"e_1_2_1_24_1","unstructured":"Free Software Foundation. 2025. GNU Operating System. https:\/\/www.gnu.org\/"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3238147.3240480"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1412.6572"},{"key":"e_1_2_1_27_1","unstructured":"Google. 2025. American Fuzzy Lop. https:\/\/github.com\/google\/AFL"},{"key":"e_1_2_1_28_1","unstructured":"gzip. 2025. The gzip home page. https:\/\/www.gzip.org\/"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3446371"},{"key":"e_1_2_1_30_1","volume-title":"Proceedings of the 33rd USENIX Security Symposium (Security '24)","author":"He Haojie","year":"2024","unstructured":"Haojie He, Xingwei Lin, Ziang Weng, Ruijie Zhao, Shuitao Gan, Libo Chen, Yuede Ji, Jiashui Wang, and Zhi Xue. 2024. Code is not Natural Language: Unlock the Power of Semantics-Oriented Graph Representation for Binary Code Similarity Detection. In Proceedings of the 33rd USENIX Security Symposium (Security '24)."},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3576915.3616605"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3243734.3243866"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/3545948.3545956"},{"key":"e_1_2_1_34_1","unstructured":"Hex-rays. 2025. IDA Pro. https:\/\/hex-rays.com\/ida-pro"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/1653662.1653736"},{"key":"e_1_2_1_36_1","volume-title":"Proceedings of the 2013 USENIX Annual Technical Conference (USENIX ATC '13).","author":"Hu Xin","year":"2013","unstructured":"Xin Hu, Kang G Shin, Sandeep Bhatkar, and Kent Griffin. 2013. MutantX-S: Scalable malware clustering based on static features. In Proceedings of the 2013 USENIX Annual Technical Conference (USENIX ATC '13)."},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3052973.3052974"},{"key":"e_1_2_1_38_1","unstructured":"Stanley Huang. 2025. MiniWeb. https:\/\/miniweb.sourceforge.net\/"},{"key":"e_1_2_1_39_1","volume-title":"Proceedings of the 22nd USENIX Security Symposium (Security '13)","author":"Jang Jiyong","year":"2013","unstructured":"Jiyong Jang, Maverick Woo, and David Brumley. 2013. Towards automatic software lineage inference. In Proceedings of the 22nd USENIX Security Symposium (Security '13)."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3597503"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2208.14191"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6311"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/SPRO.2015.10"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSE.2022.3187689"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3579856.3582823"},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-016-1941-2"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3259481"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/2897845.2897863"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1007\/11663812_11"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.3390\/app9194086"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.500"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/3510003.3510181"},{"key":"e_1_2_1_54_1","volume-title":"LIEF: Library to Instrument Executable Formats. https:\/\/lief.re\/","author":"LIEF.","year":"2025","unstructured":"LIEF. 2025. LIEF: Library to Instrument Executable Formats. https:\/\/lief.re\/"},{"key":"e_1_2_1_55_1","unstructured":"lighttpd. 2025. Lighttpd -fly light. https:\/\/www.lighttpd.net\/"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1145\/2420950.2421001"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1145\/3238147.3238199"},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00638"},{"key":"e_1_2_1_59_1","unstructured":"LLVM. 2025. The LLVM Compiler Infrastructure. https:\/\/llvm.org\/"},{"key":"e_1_2_1_60_1","unstructured":"LLVM. 2025. LLVM Discussion Forums. https:\/\/discourse.llvm.org\/"},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.5555\/3620237.3620303"},{"key":"e_1_2_1_62_1","doi-asserted-by":"publisher","DOI":"10.1145\/3433210.3453086"},{"key":"e_1_2_1_63_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1705.07874"},{"key":"e_1_2_1_64_1","doi-asserted-by":"publisher","DOI":"10.1145\/2635868.2635900"},{"key":"e_1_2_1_65_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2023.24415"},{"key":"e_1_2_1_66_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1706.06083"},{"key":"e_1_2_1_67_1","doi-asserted-by":"publisher","DOI":"10.1145\/3485447.3512045"},{"key":"e_1_2_1_68_1","volume-title":"Proceedings of the 31st USENIX Security Symposium (Security '22)","author":"Marcelli Andrea","year":"2022","unstructured":"Andrea Marcelli, Mariano Graziano, Xabier Ugarte-Pedrero, Yanick Fratantonio, Mohamad Mansouri, and Davide Balzarotti. 2022. How machine learning is solving the binary function similarity problem. In Proceedings of the 31st USENIX Security Symposium (Security '22)."},{"key":"e_1_2_1_69_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-22038-9_15"},{"key":"e_1_2_1_70_1","unstructured":"Microsoft. 2025. PE Format. https:\/\/learn.microsoft.com\/en-us\/windows\/win32\/debug\/pe-format"},{"key":"e_1_2_1_71_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1310.4546"},{"key":"e_1_2_1_72_1","unstructured":"oreans.com. 2025. Thermida Overview. https:\/\/www.oreans.com\/Themida.php"},{"key":"e_1_2_1_73_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2012.41"},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2012.08680"},{"key":"e_1_2_1_75_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2015.49"},{"key":"e_1_2_1_76_1","doi-asserted-by":"publisher","DOI":"10.1145\/2664243.2664269"},{"key":"e_1_2_1_77_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.152"},{"key":"e_1_2_1_78_1","unstructured":"NLTK Project. 2025. Natural Language Toolkit. https:\/\/www.nltk.org\/"},{"key":"e_1_2_1_79_1","unstructured":"putty. 2025. PuTTY: a free SSH and Telnet client. https:\/\/www.chiark.greenend.org.uk\/~sgtatham\/putty\/"},{"key":"e_1_2_1_80_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2022.24324"},{"key":"e_1_2_1_81_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2017.23404"},{"key":"e_1_2_1_82_1","doi-asserted-by":"publisher","DOI":"10.1145\/2133375.2133377"},{"key":"e_1_2_1_83_1","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2016.17"},{"key":"e_1_2_1_84_1","unstructured":"simoll. 2025. Remove BinaryOperator::CreateFNeg. https:\/\/reviews.llvm.org\/D75130"},{"key":"e_1_2_1_85_1","unstructured":"VMProtect Software. 2025. Advanced code security made simple and reliable. https:\/\/vmpsoft.com\/"},{"key":"e_1_2_1_86_1","doi-asserted-by":"publisher","DOI":"10.14722\/bar.2023.23005"},{"key":"e_1_2_1_87_1","unstructured":"Portable Formats Specification. 1993. Tool Interface Standard (TIS) Portable Formats Specification. (1993)."},{"key":"e_1_2_1_88_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1312.6199"},{"key":"e_1_2_1_89_1","unstructured":"GCC team. 2025. GCC the GNU Compiler Collection. https:\/\/gcc.gnu.org\/"},{"key":"e_1_2_1_90_1","unstructured":"The UPX Team. 2025. the Ultimate Packer for eXecutables. https:\/\/upx.github.io\/"},{"key":"e_1_2_1_91_1","unstructured":"tigress. 2025. the tigress c obfuscator. https:\/\/tigress.wtf\/"},{"key":"e_1_2_1_92_1","doi-asserted-by":"publisher","unstructured":"Jiyong Uhm Minseok Kim Michalis Polychronakis and Hyungjoon Koo. 2026. Fool Me If You Can: On the Robustness of Binary Code Similarity Detection Models against Semantics-preserving Transformations. doi:10.5281\/zenodo.17118200 10.5281\/zenodo.17118200","DOI":"10.5281\/zenodo.17118200"},{"key":"e_1_2_1_93_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1706.03762"},{"key":"e_1_2_1_94_1","unstructured":"VirusTotal. 2024. Welcome to YARA's documentation! https:\/\/yara.readthedocs.io\/en\/stable\/index.html"},{"key":"e_1_2_1_95_1","doi-asserted-by":"publisher","DOI":"10.1145\/3533767.3534367"},{"key":"e_1_2_1_96_1","doi-asserted-by":"publisher","DOI":"10.5555\/3698900.3699139"},{"key":"e_1_2_1_97_1","doi-asserted-by":"publisher","DOI":"10.1145\/3133956.3134018"},{"key":"e_1_2_1_98_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i01.5466"},{"key":"e_1_2_1_99_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2307.15043"},{"key":"e_1_2_1_100_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2019.23492"}],"container-title":["Proceedings of the ACM on Software Engineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3797116","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:32:54Z","timestamp":1782844374000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3797116"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,30]]},"references-count":100,"journal-issue":{"issue":"FSE","published-print":{"date-parts":[[2026,6,30]]}},"alternative-id":["10.1145\/3797116"],"URL":"https:\/\/doi.org\/10.1145\/3797116","relation":{},"ISSN":["2994-970X"],"issn-type":[{"value":"2994-970X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,30]]}}}