{
  "id": 16153,
  "url": "https://arxiv.org/abs/2608.01473v1",
  "title": "Slot2Text: Object-Centric Visual Tokenization for Efficient and Spatially Traceable Surgical MLLMs",
  "summary": "Multimodal large language models (MLLM) for surgical scene understanding typically inject hundreds of dense visual tokens into a language model, leading to costly inference and limited spatial traceability for generated answers. We present Slot2Text, a dual-mode surgical MLLM that replaces dense representations of visual input with a compact set of regions encoded as slot latents. Instead of relying on contrastive alignment of the visual encoder with language, Slot2Text groups self-supervised vi",
  "authors": "Guiqiu Liao, Matjaz Jogan, Daniel A. Hashimoto",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-02T20:09:45.000Z",
  "fetched_at": "2026-08-04T05:10:21.797Z",
  "source_slug": "arxiv-cslg",
  "source_name": "arXiv cs.LG",
  "source_homepage": "https://arxiv.org/list/cs.LG/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/16153",
  "original_url": "https://arxiv.org/abs/2608.01473v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}