About this document
Zero-Shot Stance Detection with FlanT5-XXL by SarraSaroura is a document available to read on EtoBox.
This document investigates the performance of the FlanT5-XXL model in zero-shot stance detection on tweets, demonstrating that it can match or exceed state-of-the-art benchmarks. The study explores various factors affecting performance, including prompt sensitivity, decoding strategies, and model biases. Insights from this research highlight the challenges of stance detection in social media contexts and the potential of LLMs for this task without requiring extensive labeled data.
- Author
- SarraSaroura
- Language
- EN