arXiv subjects

Harrison Green

Publications and source records attributed to Harrison Green.

5 recordsLinked to original sources

arXiv2026

Automatic, Expressive, and Scalable Fuzzing with Stitching

Fuzzing is a powerful technique for finding bugs in software libraries, but scaling it remains difficult. Automated harness generation commits to fixed API sequences at synthesis time, limiting the behaviors each harness can test. Approaches that instead explore new sequences dynamically lack the expressiveness to model real-world usage constraints leading to false positives from straightforward API misuse. We propose stitching, a technique that encodes API usage constraints in pieces that a fuzzer dynamically assembles at runtime. A static type system governs how objects flow between blocks, while a dynamically-checked extrinsic typestate tracks arbitrary metadata across blocks, enabling specifications to express rich semantic constraints such as object state dependencies and cross-function preconditions. This allows a single specification to describe an open-ended space of valid API interactions that the fuzzer explores guided by coverage feedback. We implement stitching in STITCH, using LLMs to automatically configure projects for fuzzing, synthesize a specification, triage crashes, and repair the specification itself. We evaluated STITCH against four state-of-the-art tools on 33 benchmarks, where it achieved the highest code coverage on 21 and found 30 true-positive bugs compared to 10 by all other tools combined, with substantially higher precision (70% vs. 12% for the next-best LLM-based tool). Deployed automatically on 1365 widely used open-source projects, STITCH discovered 131 new bugs across 102 projects, 73 of which have already been patched.

cs.SE↗

arXiv2025

FrameShift: Learning to Resize Fuzzer Inputs Without Breaking Them

Coverage-guided fuzzers are powerful automated bug-finding tools. They mutate program inputs, observe coverage, and save any input that hits an unexplored path for future mutation. Unfortunately, without knowledge of input formats--for example, the relationship between formats' data fields and sizes--fuzzers are prone to generate destructive frameshift mutations. These time-wasting mutations yield malformed inputs that are rejected by the target program. To avoid such breaking mutations, this paper proposes a novel, lightweight technique that preserves the structure of inputs during mutation by detecting and using relation fields. Our technique, FrameShift, is simple, fast, and does not require additional instrumentation beyond standard coverage feedback. We implement our technique in two state-of-the-art fuzzers, AFL++ and LibAFL, and perform a 12+ CPU-year fuzzer evaluation, finding that FrameShift improves the performance of the fuzzer in each configuration, sometimes increasing coverage by more than 50%. Furthermore, through a series of case studies, we show that our technique is versatile enough to find important structural relationships in a variety of formats, even generalizing beyond C/C++ targets to both Rust and Python.

cs.CR↗

arXiv2024

DiffSpec: Differential Testing with LLMs using Natural Language Specifications and Code Artifacts

Differential testing can be an effective way to find bugs in software systems with multiple implementations that conform to the same specification, like compilers, network protocol parsers, or language runtimes. Specifications for such systems are often standardized in natural language documents, like Instruction Set Architecture (ISA) specifications or IETF RFC's. Large Language Models (LLMs) have demonstrated potential in both generating tests and handling large volumes of natural language text, making them well-suited for analyzing artifacts like specification documents, bug reports, and code implementations. In this work, we leverage natural language and code artifacts to guide LLMs to generate targeted tests that highlight meaningful behavioral differences between implementations, including those corresponding to bugs. We introduce DiffSpec, a framework for generating differential tests with LLMs using prompt chaining. We demonstrate DiffSpec's efficacy on two different (extensively tested) systems, eBPF runtimes and Wasm validators. Using DiffSpec, we generated 1901 differentiating tests, uncovering at least four distinct and confirmed bugs in eBPF, including a kernel memory leak, inconsistent behavior in jump instructions, undefined behavior when using the stack pointer, and tests with infinite loops that hang the verifier in ebpf-for-windows. We also found 299 differentiating tests in Wasm validators pointing to two confirmed and fixed bugs.

cs.SE↗

arXiv2024

STRIDE: Simple Type Recognition In Decompiled Executables

Decompilers are widely used by security researchers and developers to reverse engineer executable code. While modern decompilers are adept at recovering instructions, control flow, and function boundaries, some useful information from the original source code, such as variable types and names, is lost during the compilation process. Our work aims to predict these variable types and names from the remaining information. We propose STRIDE, a lightweight technique that predicts variable names and types by matching sequences of decompiler tokens to those found in training data. We evaluate it on three benchmark datasets and find that STRIDE achieves comparable performance to state-of-the-art machine learning models for both variable retyping and renaming while being much simpler and faster. We perform a detailed comparison with two recent SOTA transformer-based models in order to understand the specific factors that make our technique effective. We implemented STRIDE in fewer than 1000 lines of Python and have open-sourced it under a permissive license at https://github.com/hgarrereyn/STRIDE.

cs.CR↗

arXiv2023

Effective Auxiliary Variables via Structured Reencoding

Extended resolution shows that auxiliary variables are very powerful in theory. However, attempts to exploit this potential in practice have had limited success. One reasonably effective method in this regard is bounded variable addition (BVA), which automatically reencodes formulas by introducing new variables and eliminating clauses, often significantly reducing formula size. We find motivating examples suggesting that the performance improvement caused by BVA stems not only from this size reduction but also from the introduction of effective auxiliary variables. Analyzing specific packing-coloring instances, we discover that BVA is fragile with respect to formula randomization, relying on variable order to break ties. With this understanding, we augment BVA with a heuristic for breaking ties in a structured way. We evaluate our new preprocessing technique, Structured BVA (SBVA), on more than 29,000 formulas from previous SAT competitions and show that it is robust to randomization. In a simulated competition setting, our implementation outperforms BVA on both randomized and original formulas, and appears to be well-suited for certain families of formulas.

cs.LO↗