Asm2Vec - Boosting Static Representation Robustness for Binary Clone Search

Explore a conference talk on Asm2Vec, an innovative approach to boosting static representation robustness for binary clone search against code obfuscation and compiler optimization. Delve into the challenges of reverse engineering and the importance of assembly clone search engines in understanding malware, finding vulnerabilities, and detecting patent infringements. Learn about the limitations of existing clone search approaches and discover how Asm2Vec jointly learns lexical semantic relationships and vector representation of assembly functions. Examine the extensive experiments conducted to benchmark this learning model against state-of-the-art static and dynamic clone search methods, and understand how the learned representation proves more robust and outperforms existing methods in the face of obfuscation and optimization changes.