Mixed campaignmathematics · combinatorics and number theory
Aletheia scans 700 open-labeled Erdős problems
A batch campaign narrowed 700 problems to 200 graded candidate outputs; 63 were technically correct but only 13 were judged meaningfully correct, while 137 were fundamentally flawed.
Editorial context
What happened
This campaign is central to the archive because it preserves a denominator and exposes failure modes that disappear when only successful cases are publicized.
Keep in view
Many technically correct outputs exploited unintended readings or rediscovered literature; the system configuration and raw outputs were not fully reproducible from the paper alone.
Chronology
Timeline
Batch runs beganattempt · day date
Batch runs endedattempt · day date
Case-study preprint postedpreprint · day date
Trust boundaries
Validation
Multiple-stage human review examined correctness, statement fidelity, and novelty.
System
AI and tools
- Aletheia
- Gemini Deep Think
People
Human contributors
- Tony Feng
- Aletheia collaborators and expert graders
Organizations
Affiliations
- Google DeepMind
Primary materials