ARC-NLP at PAN 2023: Transition-Focused Natural Language Inference for Writing Style Detection
This work addresses the problem of detecting writing style changes in multi-author documents for applications like authorship analysis, but it is incremental as it builds on existing natural language inference and Transformer methods.
The paper tackled multi-author writing style detection by framing it as a natural language inference problem using paragraph pairs, focusing on transitions between paragraphs and employing Transformer-based encoders with warmup training. The result was a model that outperformed baselines and other versions in experiments, with specific versions submitted for easy, medium, and hard setups.
The task of multi-author writing style detection aims at finding any positions of writing style change in a given text document. We formulate the task as a natural language inference problem where two consecutive paragraphs are paired. Our approach focuses on transitions between paragraphs while truncating input tokens for the task. As backbone models, we employ different Transformer-based encoders with warmup phase during training. We submit the model version that outperforms baselines and other proposed model versions in our experiments. For the easy and medium setups, we submit transition-focused natural language inference based on DeBERTa with warmup training, and the same model without transition for the hard setup.