# Typed Template Fuzzing

DevFeed: [Typed Template Fuzzing](<https://devfeed.tech/articles/typed-template-fuzzing-96661.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.31342>)

Author: Lu Maltsis, Dominik Winterer

Published: 2026-10-23T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Fuzzing/Fuzz testing](<https://devfeed.tech/topics/fuzzing.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [bugs](<https://devfeed.tech/tags/bugs.md>), [c](<https://devfeed.tech/tags/c.md>), [c-plus-plus](<https://devfeed.tech/tags/c-plus-plus.md>), [cs-pl](<https://devfeed.tech/tags/cs-pl.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [fuzzing](<https://devfeed.tech/tags/fuzzing.md>), [go](<https://devfeed.tech/tags/go.md>), [java](<https://devfeed.tech/tags/java.md>), [llms](<https://devfeed.tech/tags/llms.md>), [mutation](<https://devfeed.tech/tags/mutation.md>), [pytorch](<https://devfeed.tech/tags/pytorch.md>), [qiskit](<https://devfeed.tech/tags/qiskit.md>), [rust](<https://devfeed.tech/tags/rust.md>), [scala](<https://devfeed.tech/tags/scala.md>), [template](<https://devfeed.tech/tags/template.md>)

## AI overview

Typed Template Fuzzing (TTF) uses an LLM once in preprocessing to turn seed inputs into typed templates. Typos then performs semantically aware mutations without language-specific parsers, type checkers, or further LLM queries; correspondence between seeds and templates can also repair LLM mistakes. Evaluated across 21 input languages and libraries including PyTorch and Qiskit, Typos produced an average of 96% valid mutants. In a six-month campaign, the authors reported 135 bugs, 128 previously unknown. The abstract says Typos outperformed the state of the art, but gives no further comparison details here.

## Source excerpt

arXiv:2610.31342v1 Announce Type: new Abstract: Fuzzing software with complex inputs has been a longstanding challenge. Traditionally, fuzzers either programmatically encode language constraints, requiring significant engineering effort, or rely on users to declare grammars and constraints. Recent LLM-based fuzzers address these drawbacks but introduce new issues. Some use LLMs for test generation, sacrificing throughput and producing invalid inputs when the LLM makes mistakes. Others rely on language-specific parsers, static analysis, and AST manipulation libraries. We introduce Typed Template Fuzzing (TTF), a lightweight LLM-based fuzzing methodology. The key idea is to use LLMs in a preprocessing step to transform seeds into typed templates. The typed templates then enable semantically aware, mutation-based fuzzing without language-specific parsers, type checkers, or further LLM queries during fuzzing. By exploiting the correspondence between seeds and templates, TTF can moreover repair LLM mistakes. We have implemented TTF in our tool Typos, which we evaluate across 21 input languages, including C, C++, Go, Rust, SystemVerilog, Scala, and Java, as well as libraries such as PyTorch and Qiskit. In a six-month fuzzing campaign, we reported 135 bugs, 128 of which were previously unknown. Perhaps most notably, on average 96% of Typos's mutants are valid, coming close to the validity rates of domain-specific fuzzers. Our evaluation further shows Typos's benefits over the state of the art.