{
  "id": 18431,
  "url": "https://arxiv.org/abs/2608.10398v1",
  "title": "ELVAE: Evidential Learning-Based Variational Autoencoder for Uncertainty-Aware Generation",
  "summary": "Variational autoencoders generate samples from probabilistic latent representations but do not distinguish uncertainty about the latent location from variability around it. We formulate ELVAE, an evidential learning-based VAE in which each latent coordinate is governed by an input-dependent normal-inverse-gamma posterior. This hierarchy yields an explicit latent-location uncertainty that can be used during generation, not merely reported after inference: low-uncertainty anchors support more reli",
  "authors": "Ge Wang",
  "category": "research",
  "topics": null,
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-11T02:45:59.000Z",
  "fetched_at": "2026-08-12T05:10:43.828Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/18431",
  "original_url": "https://arxiv.org/abs/2608.10398v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}