Mutation-based Fuzzing

Technique

Mutation-based fuzzing is a fuzzing technique in which the fuzzer explores a target by mutating test artifacts such as instruction test vectors, bytestreams, or prompt templates. In the provided evidence, it is used inside coverage-guided fuzzing for RISC-V instruction-set-simulator and processor verification, as a counterexample-generation method for model learning, and as the basis of TurboFuzzLLM for black-box LLM jailbreaking-template discovery.

First seen 5/25/2026

Last seen 7/13/2026

Evidence 26 chunks

Wiki v4

WIKI

Mutation-based Fuzzing

Mutation-based fuzzing is a fuzzing technique in which exploration is driven by mutations to test artifacts. In the provided evidence, those artifacts include RISC-V instruction test vectors in coverage-guided processor verification, counterexample candidates for reactive-system model learning, and prompt templates for LLM jailbreak testing. [Mutation-based fuzzing scope]

Relationship to coverage-guided fuzzing

READ FULL ARTICLE →

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

explore full graph →

RELATIONSHIPS

14 connections

Coverage-guided Fuzzing ← uses 100% 3e

Coverage-guided fuzzing employs mutation-based techniques to generate test inputs.

hardware fuzzing implements → 90% 2e

Mutation-based fuzzing is a core technique used in hardware fuzzing.

AFL ← implements 97% 2e

AFL implements mutation-based fuzzing with bit flips, arithmetic operations, and splicing.

DiFuzzRTL ← uses 86% 2e

DifuzzRTL generates new programs through mutation-based fuzzing approaches.

Fine-Grained Code Analysis for Processor Fuzzing ← uses 90% 2e

The paper uses mutation-based fuzzing as part of its fuzz loop.

Block-Based Mutation uses → 96% 1e

Mutation-based fuzzing uses block-based mutation as one of its techniques.

Dictionary-Based Mutation uses → 96% 1e

Mutation-based fuzzing uses dictionary-based mutation as one of its techniques.

Seed Input uses → 97% 1e

Mutation-based fuzzing takes predefined seed inputs and mutates them.

Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing ← uses 95% 1e

The paper uses mutation-based algorithms in its coverage-guided fuzzing approach.

HWFuzz ← uses 100% 1e

HWFuzz uses mutation mode to modify previously generated stimuli.

TheHuzz ← uses 86% 1e

Mutations of instructions are steered by coverage signals.

AFL ← uses 97% 1e

AFL uses mutation-based input generation strategies.

Bit-Flipping Mutation uses → 96% 1e

Mutation-based fuzzing uses bit-flipping as one of its mutation techniques.

Arithmetic Mutation uses → 96% 1e

Mutation-based fuzzing uses arithmetic mutation as one of its mutation techniques.

LINKED ENTITIES

1 links

Coverage-guided Fuzzing USES Extracted graph relationship

CITATIONS

8 sources

8 citations — click to expand

[1] Mutation-based fuzzing scope Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing

[2] ISS-tailored coverage-guided mutation Verifying Instruction Set Simulators using Coverage-guided Fuzzing

[3] Fuzzing complements testcase generation Verifying Instruction Set Simulators using Coverage-guided Fuzzing

[4] Fast Exploration mutation Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing

[5] Enhanced Havoc mutation Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing

[6] Enhanced AFL results Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing

[7] Model learning counterexamples Complementing Model Learning with Mutation-Based Fuzzing

[8] TurboFuzzLLM mutation-based fuzzing TurboFuzzLLM: Turbocharging Mutation-based Fuzzing for Effectively Jailbreaking Large Language Models in Practice