ChronoCulture - Both-period accuracy - Moderate shift: leaderboard
Table 8; WVS/ISSP temporal survey pairs; model-specific top-75% knowledge filter (7767 pairs each); both predictions within 0.10 of gold; temperature 0, top-p 1, max 120 tokens; named shift stratum; FP8/AWQ checkpoints retained. Metric: Both-period accuracy (%). Higher is better. Source: https://arxiv.org/html/2610.20472v1
Metric: Both-period accuracy (%). Source: arxiv.org. Saturation forecast: Around January 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 66.4 | #78 |
| 2 | GPT-4o | 55.9 | #334 |
| 3 | Llama 3.3 70B | 47.8 | #571 |
| 4 | Qwen 3.5 35B A3B FP8 | 44.4 | #351 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=chronoculture-both-period-accuracy-moderate-shift · How It Works · Data refreshed daily, snapshot 2026-10-16.