Adversarial Learning
The philosophy behind LearnLock.
The Problem with Passive Learning
You watch a video. You read an article. You feel like you understand. But when someone asks you to explain it, you stumble. This is the illusion of competence.
Your brain confuses recognition with recall. LearnLock forces active recall by making you explain concepts, then systematically exposes what you think you know vs what you actually know.
The Pipeline
Add
YouTube, article, PDF, or GitHub URL
Extract
LLM identifies 8-12 key concepts with challenge questions
Challenge
You explain, it finds holes in your understanding
Probe
Up to 3 rounds of Socratic follow-ups
Score
1-5 based on gaps exposed, not what you got right
Schedule
SM-2 brings weak concepts back at optimal intervals
Scoring Philosophy
Unlike traditional flashcards that reward correct answers, LearnLock scores based on gaps exposed:
| Score | Meaning |
|---|---|
| 1 - Needs Work | Major gaps, fundamental misunderstanding |
| 2 - Getting There | Multiple holes exposed |
| 3 - Good | Solid base, some gaps |
| 4 - Great | Minor gaps only |
| 5 - Perfect | No holes found |
Deep dive:
Socratic Dialogue System