{"slug": "author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of", "title": "Author Representation Strategies for Zero-Shot Authorship Attribution: A Comparative Study of LLM-Based and Embedding-Based Approaches", "summary": "A comparative study of zero-shot authorship attribution (ZS AA) found that the two-stage LISA framework, which combines candidate space reduction with embedding-dimension selection, achieved the strongest overall performance among the evaluated author representation strategies. The study, posted as arXiv:2610.03531v1, evaluated a label-only prompting baseline alongside representative writing samples, LLM-generated descriptions, and style embeddings (LISA), finding label-only ZS AA ineffective while author-specific representations consistently improved attribution. LLM-generated style descriptions offered a substantially more compact representation of author style at the cost of some attribution performance, and the authors concluded current open-source LLMs remain insufficient for robust attribution without more effective representation learning.", "body_md": "arXiv:2610.03531v1 Announce Type: new \nAbstract: Authorship Attribution (AA) requires capturing fine-grained stylistic characteristics, making it particularly challenging in zero-shot (ZS) settings where no task-specific supervision is available. In this work, we investigate the effect of author representations on ZS AA by evaluating a label-only prompting baseline together with three author representation strategies: representative writing samples, LLM-generated descriptions, and style embeddings (LISA). The first three approaches perform attribution using LLM prompting, while the embedding-based approach uses style embeddings with cosine similarity. We investigate the influence of prompt design and propose a two-stage embedding-based attribution framework that combines candidate space reduction with embedding-dimension selection. The results show that label-only ZS AA is ineffective, while incorporating author-specific representations consistently improves attribution performance. Among the evaluated approaches, the proposed two-stage LISA framework achieves the strongest overall performance, whereas LLM-generated style descriptions provide a substantially more compact representation of author style at the cost of some attribution performance. These findings demonstrate the importance of author representation in ZS AA, while indicating that current open-source LLMs remain insufficient for robust attribution without more effective representation learning.", "url": "https://wpnews.pro/news/author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of", "canonical_source": "https://www.machinebrief.com/news/author-representation-strategies-for-zero-shot-authorship-at-2rav", "published_at": "2026-10-05 04:00:00+00:00", "updated_at": "2026-10-05 05:12:59.203328+00:00", "lang": "en", "topics": ["natural-language-processing", "large-language-models", "ai-research", "machine-learning", "artificial-intelligence"], "entities": ["LISA", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of", "markdown": "https://wpnews.pro/news/author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of.md", "text": "https://wpnews.pro/news/author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of.txt", "jsonld": "https://wpnews.pro/news/author-representation-strategies-for-zero-shot-authorship-attribution-a-study-of.jsonld"}}