Fusing Stylometric and Embedding Systems to Estimate Authorship Likelihood Ratios in Japanese

arXiv cs.CL Papers

Summary

This paper applies the likelihood ratio framework for forensic authorship attribution to Japanese texts, fusing stylometric features with embedding-based systems to improve discrimination and calibration.

arXiv:2606.13991v1 Announce Type: new Abstract: The likelihood ratio framework is widely recognized as the logically and legally sound basis for evidential analysis across forensic sciences, and its importance is increasingly acknowledged in analyses of authorship in textual evidence. To date, however, its application has been confined to English-language texts. Meanwhile, authorship attribution has traditionally relied on a diverse array of stylometric features, even as the rise of pre-trained large language models enables new contextual-embedding approaches. Combining these diverse approaches through fusion promises enhanced performance, yet it has not been applied to integrate stylometric-feature systems with embedding-based systems within the likelihood ratio paradigm. This study is the first to apply likelihood ratio-based forensic text comparison to Japanese digital texts, using ~1,000-character excerpts from blogs, to 1) evaluate system performance and likelihood ratio magnitudes and 2) assess the impact of fusing stylometric-feature systems with embedding-based systems. The results demonstrate that the fused system maintains excellent calibration while 1) increasing consistent-with-fact likelihood ratio magnitudes; 2) decreasing contrary-to-fact likelihood ratio magnitudes and 3) improving overall discriminability. The best-performing fusion achieved a log-likelihood-ratio cost of 0.32484, illustrating both the feasibility of likelihood ratio framework for Japanese and the benefits of fusion across heterogeneous systems.
Original Article
View Cached Full Text

Cached at: 06/15/26, 08:57 AM

# Fusing Stylometric and Embedding Systems to Estimate Authorship Likelihood Ratios in Japanese
Source: [https://arxiv.org/abs/2606.13991](https://arxiv.org/abs/2606.13991)
[View PDF](https://arxiv.org/pdf/2606.13991)

> Abstract:The likelihood ratio framework is widely recognized as the logically and legally sound basis for evidential analysis across forensic sciences, and its importance is increasingly acknowledged in analyses of authorship in textual evidence\. To date, however, its application has been confined to English\-language texts\. Meanwhile, authorship attribution has traditionally relied on a diverse array of stylometric features, even as the rise of pre\-trained large language models enables new contextual\-embedding approaches\. Combining these diverse approaches through fusion promises enhanced performance, yet it has not been applied to integrate stylometric\-feature systems with embedding\-based systems within the likelihood ratio paradigm\. This study is the first to apply likelihood ratio\-based forensic text comparison to Japanese digital texts, using ~1,000\-character excerpts from blogs, to 1\) evaluate system performance and likelihood ratio magnitudes and 2\) assess the impact of fusing stylometric\-feature systems with embedding\-based systems\. The results demonstrate that the fused system maintains excellent calibration while 1\) increasing consistent\-with\-fact likelihood ratio magnitudes; 2\) decreasing contrary\-to\-fact likelihood ratio magnitudes and 3\) improving overall discriminability\. The best\-performing fusion achieved a log\-likelihood\-ratio cost of 0\.32484, illustrating both the feasibility of likelihood ratio framework for Japanese and the benefits of fusion across heterogeneous systems\.

## Submission history

From: Shunichi Ishihara \[[view email](https://arxiv.org/show-email/8cbdba28/2606.13991)\] **\[v1\]**Fri, 12 Jun 2026 00:21:30 UTC \(1,476 KB\)

Similar Articles

Where Does Authorship Signal Emerge in Encoder-Based Language Models?

arXiv cs.CL

This paper uses mechanistic interpretability to explain why authorship attribution models fine-tuned with the same encoder, data, and loss can differ four-fold in performance depending on the scoring mechanism. It finds that the scorer determines where the encoder consolidates authorship signal, with mean pooling forcing early consolidation and late interaction allowing late consolidation.