MEME (Gold Facts) - Absence: leaderboard
Absence questions from MEME (Jung et al., 2026) answered from only the task-relevant gold facts, with no filler: recognise that a dependent value became unknown after an upstream change with no replacement rule. Accuracy in percent from a GPT-4o judge, credited only when the pre-change value was known; higher is better.
Metric: Accuracy (fraction x 100; only the task-relevant gold facts are given to the answering model, no filler and no memory system; GPT-4o judge; absence). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4 | 81 | #397 |
| 2 | Claude Opus 4.7 | 72 | #75 |
| 3 | Claude Sonnet 4.6 | 62 | #158 |
| 4 | GPT-4.1 Mini | 37 | #775 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=meme-gold-facts-absence · How It Works · Data refreshed daily, snapshot 2026-11-12.