{
  "annotation_id": "265b894d-7602-4892-9b59-4693d0aa5c0e",
  "date_created": "2026-02-17T05:53:16.008000Z",
  "date_modified": "2026-02-17T05:53:16.008000Z",
  "file_hash": "15243ad892a3fccc2d62fd7813ed836600e48c5618869272178d7dbf048a7df0",
  "private": false,
  "record": {
    "abstract": "Distracted driving is a major cause of traffic collisions, calling for robust and scalable detection methods. Vision-language models (VLMs) enable strong zero-shot image classification, but existing VLM-based distracted driver detectors often underperform in real-world conditions. We identify subject-specific appearance variations (e.g., clothing, age, and gender) as a key bottleneck: VLMs entangle these factors with behavior cues, leading to decisions driven by who the driver is rather than what the driver is doing. To address this, we propose a subject decoupling framework that extracts a driver appearance embedding and removes its influence from the image embedding prior to zero-shot classification, thereby emphasizing distraction-relevant evidence. We further orthogonalize text embeddings via metric projection onto Stiefel manifold to improve separability while staying close to the original semantics. Experiments demonstrate consistent gains over prior baselines, indicating the promise of our approach for practical road-safety applications.",
    "arxiv_id": "2601.08467",
    "authors": [
      "Takamichi Miyata",
      "Sumiko Miyata",
      "Andrew Morris"
    ],
    "categories": [
      "cs.CV",
      "cs.LG",
      "eess.IV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "Zero-Shot Distracted Driver Detection via Vision Language Models with Double Decoupling",
    "url": "https://arxiv.org/abs/2601.08467",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "82418a6f-902b-4558-9dc6-43a2c8e539fd",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}