Who Wrote the Book? Detecting and Attributing LLM Ghostwriters
This addresses the challenge of identifying LLM-generated content for applications like content moderation and academic integrity, representing a domain-specific advancement.
The paper tackles the problem of detecting and attributing authorship of long-form texts generated by frontier LLMs by introducing GhostWriteBench, a dataset with 50K+ word texts, and proposing TRACE, a novel fingerprinting method that achieves state-of-the-art performance and remains robust in out-of-distribution settings.
In this paper, we introduce GhostWriteBench, a dataset for LLM authorship attribution. It comprises long-form texts (50K+ words per book) generated by frontier LLMs, and is designed to test generalisation across multiple out-of-distribution (OOD) dimensions, including domain and unseen LLM author. We also propose TRACE -- a novel fingerprinting method that is interpretable and lightweight -- that works for both open- and closed-source models. TRACE creates the fingerprint by capturing token-level transition patterns (e.g., word rank) estimated by another lightweight language model. Experiments on GhostWriteBench demonstrate that TRACE achieves state-of-the-art performance, remains robust in OOD settings, and works well in limited training data scenarios.