Benchmarking Argumentative Behaviour of LLMs: A Study of Defences Against Character Attacks
It is argued that current safety fine-tuning constraints the strategic action space of these LLMs, making them unable to fully engage in naturalistic interactions within domains where character contestation is a normative expectation rather than a mere fallacy.
Ewelina Gajewska, Katarzyna Budzyńska, Jarosław A. Chudziak
· Frontiers in Artificial Inte... · 0 citations