Code Fingerprints: Disentangled Attribution of LLM-Generated Code

TL;DR

This paper introduces DCAN, a novel neural network approach that disentangles semantic and stylistic features to accurately attribute code snippets to their source LLMs across multiple languages.

Contribution

The paper presents the first large-scale benchmark dataset for LLM code attribution and proposes DCAN, a contrastive learning model that effectively identifies the source LLM of generated code.

Findings

DCAN achieves high attribution accuracy across models and languages.

The benchmark dataset enables systematic evaluation of code attribution methods.

Disentangling semantic and stylistic features improves attribution reliability.

Abstract

The rapid adoption of Large Language Models (LLMs) has transformed modern software development by enabling automated code generation at scale. While these systems improve productivity, they introduce new challenges for software governance, accountability, and compliance. Existing research primarily focuses on distinguishing machine-generated code from human-written code; however, many practical scenarios--such as vulnerability triage, incident investigation, and licensing audits--require identifying which LLM produced a given code snippet. In this paper, we study the problem of model-level code attribution, which aims to determine the source LLM responsible for generated code. Although attribution is challenging, differences in training data, architectures, alignment strategies, and decoding mechanisms introduce model-dependent stylistic and structural variations that serve as generative fingerprints. Leveraging this observation, we propose the Disentangled Code Attribution Network (DCAN), which separates Source-Agnostic semantic information from Source-Specific stylistic representations. Through a contrastive learning objective, DCAN isolates discriminative model-dependent signals while preserving task semantics, enabling multi-class attribution across models and programming languages. To support systematic evaluation, we construct the first large-scale benchmark dataset comprising code generated by four widely used LLMs (DeepSeek, Claude, Qwen, and ChatGPT) across four programming languages (Python, Java, C, and Go). Experimental results demonstrate that DCAN achieves reliable attribution performance across diverse settings, highlighting the feasibility of model-level provenance analysis in software engineering contexts. The dataset and implementation are publicly available at https://github.com/mtt500/DCAN.