04:00
2026-09-25
arxiv.org
large-language-models
Benchmarking Argumentative Behaviour of LLMs: A Study of Defences Against Character Attacks
A new arXiv paper (2609.28673v1) benchmarks large language models against the ElecDeb60to16-fallacy corpus of U.S. presidential debates and finds that most LLMs rigidly prioritize logical defences, fa…