MemeBridge - Sentiment: leaderboard
Sentiment labelling in MemeBridge (Zhu et al., 2026): the model labels each of 621 U.S.-originated memes positive, negative or neutral, scored against the aggregated label of four U.S. annotators. Accuracy in percent under the default prompt; higher is better.
Metric: Sentiment accuracy (%; positive, negative or neutral against the aggregated label of four U.S. annotators; 621 U.S.-originated memes with U.S. crowd labels, default prompt (no role-play), the same test given to Chinese participants). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 54.2 | #741 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=memebridge-sentiment · How It Works · Data refreshed daily, snapshot 2026-11-12.