CAVE: Controllable Authorship Verification Explanations

TL;DR

CAVE is an offline authorship verification model that provides controllable, linguistically grounded explanations, improving interpretability and maintaining competitive accuracy in privacy-sensitive applications.

Contribution

The paper introduces CAVE, a novel offline AV model that generates controllable, linguistically grounded explanations and improves interpretability over existing methods.

Findings

CAVE produces high-quality, human-evaluable explanations.

CAVE achieves competitive accuracy on AV datasets.

The approach enhances interpretability in privacy-sensitive domains.

Abstract

Authorship Verification (AV) (do two documents have the same author?) is essential in many real-life applications. AV is often used in privacy-sensitive domains that require an offline proprietary model that is deployed on premises, making publicly served online models (APIs) a suboptimal choice. Current offline AV models however have lower downstream utility due to limited accuracy (eg: traditional stylometry AV systems) and lack of accessible post-hoc explanations. In this work, we address the above challenges by developing a trained, offline model CAVE (Controllable Authorship Verification Explanations). CAVE generates free-text AV explanations that are controlled to be (1) accessible (uniform structure that can be decomposed into sub-explanations grounded to relevant linguistic features), and (2) easily verified for explanation-label consistency. We generate silver-standard training data grounded to the desirable linguistic features by a prompt-based method Prompt-CAVE. We then filter the data based on rationale-label consistency using a novel metric Cons-R-L. Finally, we fine-tune a small, offline model (Llama-3-8B) with this data to create our model CAVE. Results on three difficult AV datasets show that CAVE generates high quality explanations (as measured by automatic and human evaluation) as well as competitive task accuracy.