Skip to main content
← Choose a different target

Unlock: Hallucination Theory

Why language models confabulate, framed by the Kalai-Nachum-Vempala-Zhang (2025) statistical lower bound on pretraining error and the evaluation-incentive equilibrium that prevents post-training from removing it. Calibration, conformal prediction, structural failure modes (reversal curse, snowballing), and the measurement gap (FActScore, SAFE).

304 Prerequisites0 Mastered0 Working237 Gaps
Prerequisite mastery22%
Recommended probe

Floating-Point Arithmetic is your weakest prerequisite with available questions. You haven't been assessed on this topic yet.

Not assessed3 questions
Not assessed3 questions
Not assessed1 question
Not assessed4 questions
Ito's LemmaAdvanced
Not assessed2 questions
Not assessed1 question
Not assessed5 questions
Not assessed5 questions
Not assessed44 questions
Not assessed3 questions
Not assessed3 questions
Not assessed11 questions

Sign in to track your mastery and see personalized gap analysis.