Tag
This position paper argues that AI systems used in high-stakes decision-making should reason similarly to their users and faithfully communicate that reasoning, and outlines a research agenda for achieving such 'cognitively-aligned AI'.
This paper investigates whether five open-weight LLMs exhibit human-like sensitivity to psycholinguistic factors in anaphor resolution, using surprisal and comprehension accuracy as behavioral measures. Results show selective cognitive alignment, with some models matching human discourse sensitivity but not semantic interference effects.
Proposes Cognitive Relative Policy Optimization (CRPO), a reinforcement learning framework for aligning LLM reasoning in mental health assessment, achieving an average improvement of 10.4 percentage points in weighted F1-score over existing baselines.