Authorship Attribution

Follow this topic Ask about this topic

Authorship attribution studies how to identify a creator, decide whether two works share an author, or locate human and AI contributions within a work. Methods range from character and word n-grams and traditional stylometry to neural embeddings, contrastive learning, language-model likelihoods, and signals from handwriting, code, or narrative structure. These tools support literary and forensic analysis, privacy protection, and the detection of AI-generated or impersonated material. A central challenge is that performance can fall when genre, language, time, or model changes, while topic, editing, and mixed authorship can make stylistic evidence difficult to interpret.

Latest papers

66 papers7 papers in the last four weeks, down 30% on the four weeks before. 0.0% of all new papers.

Agent or Human? Detecting Commits by AI Coding Agents in the Absence of Authorship Signatures
Oct 26, 2026Kosei Horikawa, Hao Li, Yutaro Kashiwa +5Authorship Attribution AI Coding Agents
Nara Institute of Science and Technology, Ikoma, Japan · Queen’s University, Kingston, Canada · École de technologie supérieure - Québec University, Montréal, Canada
From Tokens to Files: Multi-Granularity Detection and Localization of AI-Involved Code
Oct 20, 2026Yue Xu, Xiyang Wang, Hui Liu +1Authorship Attribution Robustness of Machine-Generated Text Detection
Peking University · Baidu Inc.
Multi-author style detection with stylistic embeddings
Oct 16, 2026Man-Ling Sung, Chak Fai Li, Damianos Karakos +2Stylometry Authorship Attribution
RTX BBN Technologies Cambridge MA, USA
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing?
Oct 14, 2026Hasan Amin, Zhuoran Lu, Ming YinAuthorship Attribution Robustness of Machine-Generated Text Detection
Department of Computer Science, Purdue University
Who's Behind the Text? Robust Model Authorship Attribution
Oct 14, 2026Minseok Ko, Hyunseok Lee, Woomin Song +3Authorship Attribution OOD Generalization
Korea Advanced Institute of Science and Technology (KAIST)
Who Wrote It Is Not Enough: Detecting Who Contributed the Insight
Oct 5, 2026Zhuoyang Zou, Abolfazl Ansari, Jiaxi Yang +4Data Provenance Scholarly Peer Review
Pennsylvania State University, USA · University of Sheffield, United Kingdom
Writerslogic at PAN 2026: Process over Content for Robust Detection under Domain Shift
Oct 2, 2026David L. CondreyRobustness of Machine-Generated Text Detection AI-Generated Text Detection
WritersLogic Inc, San Diego, CA, USA
Author Representation Strategies for Zero-Shot Authorship Attribution: A Comparative Study of LLM-Based and Embedding-Based Approaches
Oct 2, 2026Nudrat Habib, Tosin Adewumi, Sana Sabah Al-Azzawi +2Authorship Attribution Representation Learning
Luleå University of Technology, Sweden
LA-CPD: Local-Evidence-Aware Change-Point Detection for Human-LLM Authorship Segmentation
Sep 27, 2026Qing Yang, Zhenyu Mao, Zixiang Luo +5AI-Generated Text Detection Authorship Attribution
Guilin University of Electronic Technology, China · Jinan Unaiversity, China · Jilin University, China +1
Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models
Sep 24, 2026Ehsan Barkhordar, Surendrabikram ThapaLanguage Model Fingerprinting Code Language Models
Koç University · Virginia Tech
Contrastive Learning for Authorship Verification
Sep 23, 2026Peter KirbyContrastive Learning Representation Learning
Georgia Institute of Technology
greCAPTCHA: Assessing Understanding as Evidence of Research Authorship Under Generative AI
Sep 17, 2026Justin Payan, Bálint Gyevnár, Atoosa Kasirzadeh +1Authorship Attribution Automated Evaluation
Machine Learning Department, Carnegie Mellon University · Institute for Complex Social Dynamics, Carnegie Mellon University · Departments of Philosophy & Software and Societal Systems, Carnegie Mellon University +1
Is Luke the Author of a Gospel and the Acts of the Apostles?
Sep 15, 2026Jacques SavoyAuthorship Attribution
Computer Science Department, University of Neuchâtel, Rue Emile Argand 11, 2000 Neuchâtel, Switzerland
SlopShape: Identifying AI-Generated Commercial Web Content
Sep 14, 2026Jochen MadlerAI-Generated Text Detection Language Model Fingerprinting
Sitefire
I Am No One: Style-Aware Paraphrasing for Text Anonymization
Sep 11, 2026Ahmed Sohair Khan, Estrid He, Monica Wachowicz +1Authorship Attribution Text Generation
RMIT University, Melbourne, Australia
Detecting Authorship in Political Texts with Inductive Stylometry
Sep 8, 2026Gennadii Iakovlev, Levente LittvayPolitical Discourse Analysis Stylometry
AuthBench: A Large-Scale Multilingual Benchmark for Authorship Representation across Genres and Lengths
Sep 6, 2026MaoXun Huang, Zhenxing Zhang, Claire CardieAuthorship Attribution
Department of Computer Science, Cornell University Ithaca, NY, United States
MultiGhostBench: A Multilingual Benchmark for Long-Form LLM-Generated Text Attribution under Distribution Shifts
Sep 2, 2026Matteo Greco, Anudeex Shetty, Andrea Tagarelli +1Authorship Attribution Cross-Lingual Transfer
School of Computing and Information Systems, The University of Melbourne, Australia · DIMES Dept., University of Calabria, Italy
Conducting Stylistic Analysis of Paintings through an Art-History Agent
Aug 30, 2026Marc S. Walton, Astrid HarthExplainable Artificial Intelligence Stylometry
Museum Studies Programme, The University of Hong Kong, Hong Kong SAR · Department of Chinese and History, City University Hong Kong, Hong Kong SAR
Aggregation-Aware Synthetic Text Generation Against Authorship Re-Identification
Aug 23, 2026Qian Ma, Anna Squicciarini, Sarah RajtmajerText Generation Authorship Attribution
Information Sciences and Technology The Pennsylvania State University
PersonalBench: Measuring the Authorship Gap in LLM Personalization
Aug 20, 2026Yash Ganpat SawantPersonalized Language Models Personalized Text Generation
Independent AI Researcher
When Writing Style Drifts: Benchmarking Authorship Verification under Distribution Shifts in Genre, Time and the AI-Era
Aug 18, 2026Lotta Kiefer, Brisca Balthes, Christoph Leiter +3Distribution Shift Robustness Authorship Attribution
University of Technology Nuremberg (UTN)
Writing Style Similarity Reflects Academic Genealogy
Aug 14, 2026Cameron ManzoStylometry Authorship Attribution
RTX BBN Technologies
Stylometric Defenses Against Author Impersonation in Software Repositories
Aug 3, 2026Leonid Ravich, Michael FireOnline Anomaly Detection Stylometry
Ben-Gurion University of the Negev, Computer Science Department
IDraw: Artist Verification from Digital Drawing Images
Aug 3, 2026Nayoung Kim, Nan Jiang, Bangjie Sun +3Digital Image Forensics Stylometry
1Korea Advanced Institute of Science and Technology (KAIST) · 2National University of Singapore (NUS)
Authorship Verification of Transcribed German-Language Videos
Jul 31, 2026Oren Halvani, Sophie TitzeStylometry Authorship Attribution
Fraunhofer Institute for Secure Information Technology SIT Darmstadt, Germany
Theatre Chapbooks At Scale: A Statistical Comparative Analysis of Typography
Jul 29, 2026Diego Belzarena, Seginus Mowlavi, Paula Casariego Castiñeira +3Historical Document Analysis Computational Literary Studies
UDELAR, CB · Université Paris-Saclay, ENS Paris-Saclay, Centre Borelli, France · CB +8
CHiPS: Character Histograms and Positional Signals for Lightweight Authorship Attribution in Romanian Texts
Jul 24, 2026Sanda-Maria Avram, George C. ŢurcaşText Classification Stylometry
Department of Computer Science, Faculty of Mathematics and Computer Science, Babeș-Bolyai University, Cluj-Napoca, Romania · Department of Mathematics, Faculty of Mathematics and Computer Science, Babeș-Bolyai University, Cluj-Napoca, Romania
Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution
Jul 16, 2026Zlata Kikteva, Artur Romazanov, Annette Hautli-Janisz +1AI-Generated Text Detection Authorship Attribution
University of Passau, Germany · University of Dundee, United Kingdom
Normalisation-Based Likelihood Ratio Estimation for Forensic Authorship Verification
Jul 10, 2026Sadie Barlow, Andrea Nini, Edoardo ManinoDigital Forensics Authorship Attribution
The University of Manchester, Department of Linguistics and English Language, Oxford Road, Manchester, M13 9PL · The University of Manchester, Department of Computer Science, Oxford Road, Manchester, M13 9PL

Oct 26, 2026J/K move · Enter open · S save

Agent or Human? Detecting Commits by AI Coding Agents in the Absence of Authorship Signatures

Kosei Horikawa, Hao Li, Yutaro Kashiwa, Youssef Esseddiq Ouatiti+4

Nara Institute of Science and Technology, Ikoma, Japan · Queen’s University, Kingston, Canada · École de technologie supérieure - Québec University, Montréal, Canada

Authorship Attribution AI Coding Agents Software Engineering Agents

Open paper AskPDF arXiv

AI coding agents such as Claude Code, GitHub Copilot coding agent, and Cursor now commit directly to open-source repositories. Empirical studies of these agents identify their commits through authorship signatures, such as bot accounts and Co-authored-by trailers. These signatures are optional, can be removed, and are often attached to pull requests rather than to individual commits. Hence, it is unclear whether Agent commits can be told apart from Human commits once the signatures are gone. We investigate the signals and limits of signature-free detection using 7,837,946 commits from 1,226 repositories, covering five coding agents and Human commits. After signatures are masked, detectors rank Agent commits above Human commits with a ROC AUC of up to 0.859, and most of this signal comes from message verbosity and change size rather than from the code itself. On the shared sample, gpt-6-sol without fine-tuning reaches AUCs similar to XGBoost and TF-IDF. At the observed Agent share, however, the best detector on the full test set reaches a precision of only 0.332. Detection transfers poorly to held-out Claude Code and Copilot, with TF-IDF reaching an AUC of only 0.524 for Claude Code. Temporal changes also depend on the agent, e.g., Copilot AUC drops and later recovers, while more training history helps Claude Code but not Copilot. These results indicate that signature-free detection can prioritize commits for review but cannot replace recorded attribution, and that studies relying on signature-based labels should treat them as a lower bound on agent activity.