Learning to Fuzz: Application-Independent Fuzz Testing with Probabilistic, Generative Models of Input Data

By Jibesh Patra, Michael Pradel. 2016

65 citations recorded by the source corpus.

How researchers see this paper

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

In Natural Language Processing

Citations in the corpus

A Survey of Machine Learning for Big Code and Naturalness (2017, methodology): “alize the code by consistentlyalpha-renamingvariablesandmethodnames.Finally,theyﬁlterinvalidintermediateoutput so they, in the end, generate only valid programs. In a similar manner, Patra and Pradel [153] synthesize JavaScript programs for fuzz testing JavaScript interpreters. 5.8 Program Analysis ”
DeepBugs: a learning approach to name-based bug detection (2018, methodology): “…predicting fixes of compilation errors via neural networks [Bhatia and Singh 2016; Gupta et al. 2017], and the generation of inputs for fuzz testing via probabilistic, generative language models [Patra and Pradel 2016] or neural networks [Amodio et al. 2017; Godefroid et al. 2017; Liu et al. 2017].”
AFLNET: A Greybox Fuzzer for Network Protocols (2020, background): “We can distinguish blackbox approaches [19], [20] that learn the message structure from a given corpus of messages and whitebox approaches [21], [22] that actively explore the protocol implementation to uncover message structure.”
Context2Name: A Deep Learning-Based Approach to Infer Natural Variable Names from Usage Contexts (2017, background): “to programs, targeting a variety of development tasks such as code completion [10, 29, 32], fixing syntactic errors [9, 18], generating inputs for fuzz testing of programs [17, 24, 28], and generating programs or short code snippets to assist developers or test compilers [4, 34].”
Superion: Grammar-Aware Greybox Fuzzing (2018, unknown): “r to generate C programs. Domato [21] generates test inputs from scratch given the grammars that specify HTML/CSS structures and JavaScript objects, properties and functions. Skyﬁre [60] and TreeFuzz [47] learns a probabilistic model, using the grammar and a corpus of test inputs, to generate new te”
Learn&Fuzz: Machine learning for input fuzzing (2017, background): “TreeFuzz [24] was recently proposed as a fuzz testing approach for tree structured inputs (such as programs) by learning a generative model of tree structures from a corpus of example data.”
Improving bug detection via context-based code representation learning and attention-based neural networks (2019, methodology): “…engineering (SE) tasks, such as automated correction for syntax errors [Bhatia and Singh 2016], fuzz testing with probabilistic, generative models [Patra and Pradel 2016], program synthesis [Amodio et al. 2017], code clones [Li et al. 2017; Smith and Horwitz 2009; White et al. 2016], program…”
NL2Type: Inferring JavaScript Function Types from Natural Language Information (2019, methodology): “Other work uses neural networks to predict parts of code [29], to detect vulnerabilities [30], to generate inputs for fuzz testing [56], [57], to detect and ﬁx syntactic programming mistakes [58], and to predict whether a ﬁle is likely to contain a bug [59]. g) JavaScript Analysis: Program analyses…”
Fuzz4ALL: Universal Fuzzing with Large Language Models (2023, background): “TreeFuzz [60] parses the training corpus into a tree structure and through tree traversal, learns a probabilistic, generative model that synthesizes new fuzzing inputs.”
Large Language Model guided Protocol Fuzzing (2024, unknown): “Blackbox fuzzers such as T REE F UZZ [35] employ machine learning techniques over the seed corpus to construct probabilistic models that are subsequently used for input generation.”
Deep Learning to Find Bugs (2017, background): “Existing approaches address a variety of development tasks, including code completion based on probabilistic models of code [9, 34, 36], predicting fixes of compilation errors via neural networks [8, 14], and the generation of inputs for fuzz testing via probabilistic, generative language models [30”