Different scoring mechanisms cause encoder-based authorship attribution models to consolidate authorship signals at different layers, as shown by causal interventions and gradient analysis.
Isolating Authorship from Content with Semantic Embeddings and Contrastive Learning , url =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2representative citing papers
Short LLM prompts contain distinctive lexical signals enabling user identification as a behavioral biometric, with lexical features outperforming semantic ones across a dataset of 20k+ prompts from 1k users.
citing papers explorer
-
Where Does Authorship Signal Emerge in Encoder-Based Language Models?
Different scoring mechanisms cause encoder-based authorship attribution models to consolidate authorship signals at different layers, as shown by causal interventions and gradient analysis.