Justin Shenk

Legal LLMs trained on surface features like citation count and legalese density learn to strategically withhold answers, becoming maximally "lawyerly" but minimally useful.

Steering a language model's intertemporal preferences can induce significant shifts in decision-making, impacting how AI systems advise on delayed costs and benefits.

Interpolating between opposing directorial personas not only reveals surprising coherence improvements but also uncovers a shared moral-tone substrate in transformer models.