{
  "id": 11346477,
  "title": "Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes",
  "url": "https://urgent.news/2026/10/01/where-opd-spatially-guided-on-policy-self-distillation-of-mllms-with",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-01T17:34:10.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.02117v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a frozen or EMA version of themselves that receives privileged information. Its application to multimodal large language models (MLLMs), however, remains largely unexplored. Recent approaches use privileged visual information, such as image crops…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}