ContrAR (Single Frame) - Smart Retail: 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 middle frame of each video. Detection accuracy in percent on the smart retail videos; higher is better.

Metric: Detection accuracy (%) for contradictory virtual content attacks in AR videos (66 smart retail videos; half the videos carry a contradiction, a balanced yes/no decision), one middle frame of each video given (single-frame prompting); mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-586.36#178
2Gemini 2.5 Pro81.82#276
3GPT-4o75.76#751
4GPT-4.171.21#505
5Gemini 2.5 Flash69.7#463
6Grok 468.18#319
7Grok 2 Vision63.64#1437
8Claude Sonnet 4.560.61#261
9Qwen 2.5 VL 72B Instruct59.09#841
10Claude Haiku 4.556.06#583
11Qwen 2.5 VL 7B Instruct56.06#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-single-frame-smart-retail · How It Works · Data refreshed daily, snapshot 2026-11-21.