{
  "absUrl": "https://arxiv.org/abs/2610.12081v1",
  "arxivId": "2610.12081",
  "arxivVersion": 1,
  "arxivVersionedId": "2610.12081v1",
  "authors": [
    {
      "affiliations": [
        "Google DeepMind"
      ],
      "name": "Fedor Kitashov"
    },
    {
      "affiliations": [
        "Google DeepMind"
      ],
      "name": "João Carreira"
    },
    {
      "affiliations": [
        "Toyota Technological Institute at Chicago"
      ],
      "name": "Shiry Ginosar"
    },
    {
      "affiliations": [
        "Google DeepMind",
        "University of Bristol"
      ],
      "name": "Dima Damen"
    },
    {
      "affiliations": [
        "Google DeepMind",
        "University of Oxford"
      ],
      "name": "Andrew Zisserman"
    },
    {
      "affiliations": [
        "Google DeepMind"
      ],
      "name": "Viorica Pătrăucean"
    }
  ],
  "contract": "researcher-sidecars-v1",
  "id": "2610.12081v1",
  "originalTitle": "Perception Test 2026: Challenge Summary and Extension to City-scale Audio-Visual Reasoning",
  "pdfUrl": "https://arxiv.org/pdf/2610.12081v1.pdf",
  "schemaVersion": 1,
  "type": "preprint"
}