{
  "id": 15860,
  "url": "https://arxiv.org/abs/2607.28590",
  "title": "VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation",
  "summary": "Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how strongly to distill. We introduce Visual Attribution Distillation (VAD), a counterfactual target-reconstru",
  "authors": "Kangning Zhang, Yixing Li, Shuai Shao, Qingyao Li, Zhengxi Lu, Zhiyuan Yao",
  "category": "research",
  "topics": "regulation,children-education",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-29T20:00:00.000Z",
  "fetched_at": "2026-08-04T05:10:21.797Z",
  "source_slug": "hf-daily",
  "source_name": "HuggingFace Daily Papers",
  "source_homepage": "https://huggingface.co/papers",
  "ethics_ai_record_url": "https://ethics.ai/record/15860",
  "original_url": "https://arxiv.org/abs/2607.28590",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}