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).
407 Prerequisites0 Mastered0 Working274 Gaps
Prerequisite mastery33%
Recommended probe
Peano Axioms is your weakest prerequisite with available questions. You haven't been assessed on this topic yet.
Hallucination TheoryTARGET
Not assessed5 questions
Chain-of-Thought and ReasoningFrontier
Not assessed3 questions
Not assessed44 questions
Not assessed3 questions
Context EngineeringFrontier
Not assessed3 questions
Not assessed3 questions
Multimodal RAGFrontier
No quiz
Out-of-Distribution DetectionAdvanced
No quiz
No quiz
RLHF and AlignmentResearch
Not assessed3 questions
Transformer ArchitectureResearch
Not assessed11 questions
Agent Protocols: MCP and A2AFrontier
No quiz
Sign in to track your mastery and see personalized gap analysis.