{
  "id": 5898,
  "url": "https://arxiv.org/abs/2604.12223v1",
  "title": "LLM-Guided Semantic Bootstrapping for Interpretable Text Classification with Tsetlin Machines",
  "summary": "Pretrained language models (PLMs) like BERT provide strong semantic representations but are costly and opaque, while symbolic models such as the Tsetlin Machine (TM) offer transparency but lack semantic generalization. We propose a semantic bootstrapping framework that transfers LLM knowledge into symbolic form, combining interpretability with semantic capacity. Given a class label, an LLM generates sub-intents that guide synthetic data creation through a three-stage curriculum (seed, core, enri",
  "authors": "Jiechao Gao, Rohan Kumar Yadav, Yuangang Li, Yuandong Pan, Jie Wang, Ying Liu et al.",
  "category": "research",
  "topics": "safety-alignment,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-14T03:02:25.000Z",
  "fetched_at": "2026-07-14T16:32:06.467Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/5898",
  "original_url": "https://arxiv.org/abs/2604.12223v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}