ContrAR (Multiple Frames) - Smart Apartment: leaderboard

ContrAR (2026) tests whether vision-language models detect contradictory virtual content attacks in augmented reality, answering yes or no on real AR videos of which half carry an overlay that contradicts the physical scene, given the first, middle and last frames of each video. Detection accuracy in percent on the smart apartment videos; higher is better.

Metric: Detection accuracy (%) for contradictory virtual content attacks in AR videos (90 smart apartment videos; half the videos carry a contradiction, a balanced yes/no decision), first, middle and last frames given (multi-frame prompting); mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-4.193.33#505
2GPT-592.22#178
3GPT-4o86.67#751
4Gemini 2.5 Pro82.22#276
5Gemini 2.5 Flash76.67#463
6Claude Sonnet 4.567.78#261
7Qwen 2.5 VL 72B Instruct58.89#841
8Grok 457.78#319
9Grok 2 Vision56.67#1437
10Claude Haiku 4.555.56#583
11Qwen 2.5 VL 7B Instruct54.44#1902

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=contrar-multiple-frames-smart-apartment · How It Works · Data refreshed daily, snapshot 2026-11-21.