Review4Repair: Code review aided automatic program repairing

Huq, Faria; Hasan, Masum; Pantho, Mahim Anzum Haque; Mahbub, Sazan; Iqbal, Anindya; Ahmed, Toufique

doi:10.1016/j.infsof.2021.106765

Cited by 8 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine Learning for Program Analysis. Machine learning has shown great promises in analyzing both source code and executables [3,24,62,74] in tasks like type inference [37,56,61,64,70,92], code completion [10,17,43], program synthesis and generation [79,88], program repair and fix [2,26,41,80,97], code summarization [14,21,75,84], general code representation [13,39,46,51,53,89], bug/vulnerability detection [23,48,63,73], code clone detection and search [14,32,33,42,55,66], code translation [72], comment suggestion [40,52], and reverse engineering tasks [7,9,45,65]. Recent works have observed that incorporating prog...…”

Section: Related Workmentioning

confidence: 99%

NeuDep: neural binary memory dependence analysis

Pei

She

Wang

et al. 2022

Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Enginee

View full text Add to dashboard Cite

Determining whether multiple instructions can access the same memory location is a critical task in binary analysis. It is challenging as statically computing precise alias information is undecidable in theory. The problem aggravates at the binary level due to the presence of compiler optimizations and the absence of symbols and types. Existing approaches either produce significant spurious dependencies due to conservative analysis or scale poorly to complex binaries.We present a new machine-learning-based approach to predict memory dependencies by exploiting the model's learned knowledge about how binary programs execute. Our approach features (i) a self-supervised procedure that pretrains a neural net to reason over binary code and its dynamic value flows through memory addresses, followed by (ii) supervised finetuning to infer the memory dependencies statically. To facilitate efficient learning, we develop dedicated neural architectures to encode the heterogeneous inputs (i.e., code, data values, and memory addresses from traces) with specific modules and fuse them with a composition learning strategy.We implement our approach in NeuDep and evaluate it on 41 popular software projects compiled by 2 compilers, 4 optimizations, and 4 obfuscation passes. We demonstrate that NeuDep is more precise (1.5×) and faster (3.5×) than the current state-of-the-art. Extensive probing studies on security-critical reverse engineering tasks suggest that NeuDep understands memory access patterns, learns function signatures, and is able to match indirect calls. All

show abstract