Reproduce “Unreliable Narrator Framing as a Cost-effective Mechanism for Satire Explanation of Australian and American Satirical News” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Kanay Gupta, David Y. Liu, Aditya Joshi
The authors test whether a prompt separating satire’s straight-faced account from its implied meaning improves model explanations of Betoota Advocate and The Onion articles. It asks models to identify a narrator trope, state the surface account, infer the author’s meaning, and explain their difference. Llama-3.1-8B-Instruct’s scores rose from 2.83 to 3.57 out of 5 on Betoota; target grounding remained weak.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.