{
  "url": "https://appliedfrontier.nl/en/research/signal-3e9b69b6580d0429/",
  "title": "Process-Verified Reinforcement Learning for Theorem Proving via Lean",
  "published": "2026-06-18",
  "excerpt": "While reinforcement learning from verifiable rewards (RLVR) typically has relied on a single binary verification signal, symbolic proof assistants in formal reasoning offer rich, fine-grained",
  "content_sha256": "4bd378a25ca5cc0fe921a6b3e06fd584cd475436bb3f696d6736d19d36421ce8",
  "extracted_chars": 1538,
  "artifact_links": [
    "https://arxiv.org/abs/2606.20068"
  ],
  "arxiv_id": "2606.20068",
  "id": "signal-3e9b69b6580d0429",
  "org": "Research community",
  "topic": "Formal verification",
  "source_kind": "preprint",
  "status": "unreviewed-source-report",
  "analysis_method": "deterministic_metadata_triage_v1",
  "novelty_verified": false,
  "proof_checked": false,
  "peer_review_verified": false,
  "matched_terms": [
    "lean",
    "proof",
    "reasoning",
    "theorem"
  ],
  "checked_at": "2026-10-29T22:08:08+00:00",
  "first_seen": "2026-10-29T21:52:00+00:00",
  "source_hash": "938255f08eafbe4d0ffd7d57e6f922ec361b9ccf955699021bffb2058c291d22",
  "source_url": "https://arxiv.org/abs/2606.20068",
  "language": "en",
  "indexing": "noindex_until_editorially_enriched"
}