cs.CLNov 4, 2026

Consistent yet Contextual: Learning Author-Level Mixtures of Style for Authorship Attribution

Authors: Omkar Yadav, Michael Jiang, Kenan Alkiek, Jian Zhu, David Jurgens

Organizations: University of Michigan · University of British Columbia

Abstract

Approaches to authorship verification or attribution typically model the style of a document, asking whether two documents were likely written by the same author. While recent approaches have been performant, such approaches largely ignore modeling the authors themselves and how varied their style is. Here, we introduce multiple alternatives for modeling the author and demonstrate that allowing an author's representation to capture multiple consistent but distinct writing styles yields significant performance improvements. Using our new authorship model, we then examine when an author writes in a way that is most idiomatic of their style across four scenarios: (i) variation in the size of their writing audiences, (ii) temporal distance from when the model was trained, (iii) the effect of preceding context, and (iv) the effect of genre norms on style. Across all four experiments, we demonstrate that authors vary substantially in how idiolectic their writing is. Our results have clear implications for both the modeling and data used for authorship verification and attribution, showing that authors need to be represented with their multiple styles and that some settings produce text that is much more informative about who wrote that text.

Figures & tables

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Oct 15, 2025

One-shot Style Transfer LLM log-probabilities for Authorship Attribution and Verification

Computational stylometry studies writing style through quantitative textual patterns, enabling applications such as authorship attribution, identity linking, and plagiarism detection. Despite the relevance of language modeling to these tasks, the pre-training of modern large language models (LLMs) has been underutilized in authorship attribution and verification. We introduce an unsupervised framework that uses the log-probabilities of an LLM to measure style transferability between two texts. This framework takes advantage of the extensive Causal Language Modeling (CLM) pre-training, one-shot capabilities and scale of LLMs, avoiding explicit supervision. Our methods substantially outperform prompting-based unsupervised baselines in authorship verification at similar model sizes, and is competitive with or improves contrastive baselines in most settings with sufficient model scale. We further observe strong performance across non-English languages. The effectiveness of the proposed framework improves consistently with increasing model scale. In the case of authorship verification, we propose an additional mechanism that increases test-time computation to improve accuracy; enabling flexible trade-offs between computational cost and task performance.
Sep 6, 2026

AuthBench: A Large-Scale Multilingual Benchmark for Authorship Representation across Genres and Lengths

Authorship signals matter in settings where writing style carries identity: digital forensics, plagiarism analysis, account linking, misinformation investigation, and machine-generated text detection. Yet current authorship benchmarks remain fragmented, usually covering only a narrow language set, a single genre, or a limited document-length regime, which makes it difficult to assess whether modern representations truly generalize. We introduce AuthBench, a large-scale multilingual benchmark for authorship representation that is designed to make this evaluation broad, standardized, and realistic. AuthBench contains 428,150 documents written by 153,825 individuals across ten widely used languages, 9 primary genres, 66 fine-grained genres, and four document-length buckets. It supports two complementary tasks: authorship attribution, formulated as same-author retrieval and authorship verification, formulated as same-author binary decision. We benchmark 47 neural models and three non-neural baselines under a unified zero-shot protocol. Results show that authorship representation remains far from solved: the best retrieval model reaches only 0.258 Success@5, while the best verification model achieves 0.076 EER and 0.968 ROC-AUC. The leaderboard also reveals a meaningful task split, with different model families leading retrieval and verification, and large performance differences across languages, genres, and lengths. These findings position AuthBench not only as a new benchmark, but as a diagnostic resource for studying when and why authorship representations succeed or fail. We release AuthBench, its evaluation toolkit, and benchmark data at https://github.com/mao-code/AuthBench and https://huggingface.co/datasets/MaoXun/AuthBench.
Aug 18, 2026

When Writing Style Drifts: Benchmarking Authorship Verification under Distribution Shifts in Genre, Time and the AI-Era

Authorship verification (AV) assumes that an author's writing style remains sufficiently stable to distinguish it from that of other writers. In practice, however, this assumption is challenged by distribution shifts caused by changes in genre, time, and AI-assisted writing. Existing AV benchmarks typically study these factors in isolation and focus predominantly on English, limiting our understanding of model robustness under realistic conditions. We introduce AVShift, the first German benchmark for systematically evaluating AV under multiple distribution shifts. AVShift comprises over 150,000 text pairs spanning three genres and 21 years, enabling controlled evaluation of cross-genre, temporal, and AI-era shifts within a unified framework. We benchmark representative feature-based, embedding-based, and LLM-based approaches. Our experiments show that fine-tuned LLMs generalize best across genres and benefit substantially from stylistically diverse training data. We further demonstrate that temporal drift is one of the strongest factors affecting AV, with performance degrading significantly as the time gap between documents increases. In contrast, we find no evidence of a measurable AI-era distribution shift within AVShift. Finally, our feature analysis reveals stylistic features that remain stable across genres, while their relative importance varies depending on the specific genre transition. We release AVShift and our code for future research.