Statistical Deobfuscation of Android Applications

By Benjamin Bichsel, Veselin Raychev, Petar Tsankov, Martin T. Vechev. 2016

133 citations recorded by the source corpus.

Subfields: Information Systems, Software, Program Analysis & Verification, Privacy & Security, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

1. Miltiadis Allamanis
2. Charles Sutton
3. Uri Alon
4. Eran Yahav
5. Earl T. Barr
6. Omer Levy
7. Marc Brockschmidt
8. Vincent J. Hellendoorn

In Software

1. Charles Sutton
2. Miltiadis Allamanis
3. Earl T. Barr
4. Premkumar T. Devanbu
5. Vincent J. Hellendoorn
6. Marc Brockschmidt
7. Uri Alon
8. Eran Yahav

In Program Analysis & Verification

1. Miltiadis Allamanis
2. Charles Sutton
3. Eran Yahav
4. Uri Alon
5. Marc Brockschmidt
6. Omer Levy
7. Earl T. Barr
8. Premkumar T. Devanbu

In Privacy & Security

1. Marc Brockschmidt
2. Miltiadis Allamanis

In Signal Processing

1. Uri Alon
2. Eran Yahav
3. Miltiadis Allamanis
4. Charles Sutton
5. Omer Levy
6. Marc Brockschmidt
7. Earl T. Barr
8. Patrick Fernandes

Citations in the corpus

• A Survey of Machine Learning for Big Code and Naturalness (2017, background): “[27] Dependency Net Identifier Name CRF (GM) Deobfuscation”
• Learning to Represent Programs with Graphs (2017, methodology): “Raychev et al. (2015) and Bichsel et al. (2016) use conditional random fields to model a variety of relationships between variables, AST elements and types to predict variable names and types (resp. to deobfuscate Android apps), but without considering the flow of data explicitly.”
• code2vec: learning distributed representations of code (2018, methodology): “Generalization ability Using CRFs for predicting program properties was found to be powerful [Bichsel et al. 2016; Raychev et al. 2015], but limited to modeling only combinations of values that were seen in the training data.”
• DOBF: A Deobfuscation Pre-Training Objective for Programming Languages (2021, background): “, 2014; 2015), probabilistic models (Raychev et al., 2015; Bichsel et al., 2016; Vasilescu et al., 2017; Alon et al., 2018), and recurrent neural networks (Bavishi et al.”
• Language-Agnostic Representation Learning of Source Code from Structure and Context (2021, background): “These are useful for deobfuscating Android applications (Bichsel et al., 2016) for example.”
• Structured Neural Summarization (2018, methodology): “Raychev et al. (2015) and Bichsel et al. (2016) use CRFs for a range of tasks on source code, including the inference of names for variables and methods.”
• Deep learning type inference (2018, background): “Other applications include deobfuscation [10], coding conventions [7, 28] and migration [21, 25].”
• Generative Code Modeling with Graphs (2018, methodology): “More recently, methods handling both modalities ( i.e. , the formal and natural language aspects) have shown successes on important software engineering tasks [3, 5, 21].”