{
  "id": 14054,
  "url": "https://arxiv.org/abs/2607.24354v1",
  "title": "Are Prompt Optimizers Blind? Cross-Modal Visual Feedback for Automatic Prompt Optimization",
  "summary": "Automatic prompt optimization (APO) has been widely adopted to adapt vision-language models (VLMs) to downstream tasks without weight updates, yielding promising results. However, on multimodal tasks, the effectiveness of APO is fundamentally bottlenecked by a blind feedback channel: the optimizer reads the question, the prediction, and the gold answer, but never the input image on which the model failed, and therefore cannot diagnose visually grounded errors. As a remedy, we introduce Cross-Mod",
  "authors": "Haoyue Liu, Xiaoyu Ma, Ye Chen, Yuexian Zou, Xiaoying Tang",
  "category": "research",
  "topics": "healthcare",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-27T12:31:58.000Z",
  "fetched_at": "2026-07-28T05:10:12.325Z",
  "source_slug": "x-arxiv-cs-ai",
  "source_name": "arXiv cs.AI",
  "source_homepage": "https://arxiv.org/list/cs.AI/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/14054",
  "original_url": "https://arxiv.org/abs/2607.24354v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}