Tag
This paper investigates how LLMs rely on morphological cues (affixes) to make pharmacological inferences, demonstrating that models can confidently generate plausible content for fictitious drug names based solely on affix heuristics, which poses a subtle safety risk.
This paper investigates mismatches between expected and actual time savings when using LLMs for simple cognitive tasks, revealing a speedup illusion where users underestimate AI-assisted completion times despite no actual speedup.
This study analyzes how personality, model backbone, and guardrails influence the emergent social behavior of deployed AI agents on the Moltbook platform, finding that personality specification is the dominant factor.