{
  "id": 16980,
  "url": "https://arxiv.org/abs/2608.04084v1",
  "title": "SpecDrop: Parameter-Free Category-Conditioned Routing for Modular Specialization",
  "summary": "Mixture-of-experts (MoE) networks pursue specialization through learned routers, gates, and load-balancing losses, yet at matched total-parameter budgets learned routers can underperform equal-weight No-Routing baselines. Is the bottleneck the routing algorithm, or the alignment between training-signal granularity and the target categories? We probe the question with SpecDrop, a fixed parameter-free routing scheme: each of $K$ branches receives weight $p_a$ for its assigned category and a small",
  "authors": "Boyao Wang, Zhihan Lei",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-04T18:00:01.000Z",
  "fetched_at": "2026-08-06T05:10:11.148Z",
  "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/16980",
  "original_url": "https://arxiv.org/abs/2608.04084v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}