{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LYA2NSAFTESK7RUQLR2WERHYEZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e63abfc126789fbe45cdc0b9e889da8c76f9adfe07e8ec7b4c3bf593e3a32064","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-25T13:03:09Z","title_canon_sha256":"4be6f49c9d013e4b4bc84e95f17e73b37d2418715feb236f33c05958f9c6f933"},"schema_version":"1.0","source":{"id":"2503.19618","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.19618","created_at":"2026-07-05T11:11:21Z"},{"alias_kind":"arxiv_version","alias_value":"2503.19618v2","created_at":"2026-07-05T11:11:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19618","created_at":"2026-07-05T11:11:21Z"},{"alias_kind":"pith_short_12","alias_value":"LYA2NSAFTESK","created_at":"2026-07-05T11:11:21Z"},{"alias_kind":"pith_short_16","alias_value":"LYA2NSAFTESK7RUQ","created_at":"2026-07-05T11:11:21Z"},{"alias_kind":"pith_short_8","alias_value":"LYA2NSAF","created_at":"2026-07-05T11:11:21Z"}],"graph_snapshots":[{"event_id":"sha256:e2e40cc872a6c1dde2beda3c2151aeff57836cc830379e7c1b779c74af83e96a","target":"graph","created_at":"2026-07-05T11:11:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.19618/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose to scale RL to unverifiable data with a novel algorithm JEPO (Jensen's Evidence lower bound Policy Optimization). While most prior efforts on scaling RL for LLMs focus on verifiable data where ground truth answers are typically short-form and can be matched easily; we investigate the case where such assumptions are less valid (e.g., when answers are long-form such as mathematical proofs). To scale RL training to unverifiable data with contemporary training constraints, we propose JEPO. JEPO applies Jensen's evidence lower bound, a pragmatic simplification of the evidence lower bound","authors_text":"Lovish Madaan, R\\'emi Munos, Sid Wang, Yunhao Tang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-25T13:03:09Z","title":"Beyond Verifiable Rewards: Scaling Reinforcement Learning for Language Models to Unverifiable Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19618","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3cd8511672405d14ecc171279e9124c786fe5a99a5e92979d33d31327cc6c5ce","target":"record","created_at":"2026-07-05T11:11:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e63abfc126789fbe45cdc0b9e889da8c76f9adfe07e8ec7b4c3bf593e3a32064","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-25T13:03:09Z","title_canon_sha256":"4be6f49c9d013e4b4bc84e95f17e73b37d2418715feb236f33c05958f9c6f933"},"schema_version":"1.0","source":{"id":"2503.19618","kind":"arxiv","version":2}},"canonical_sha256":"5e01a6c8059924afc6905c756244f8265bfa57fb3422afdff9c669a712111865","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5e01a6c8059924afc6905c756244f8265bfa57fb3422afdff9c669a712111865","first_computed_at":"2026-07-05T11:11:21.903985Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:11:21.903985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZOdME6fsJeOtqOpgDdFIlWANR281Z4d1T/6UG3fzAjBKGSvZHQSTmXP8cNaGEmBAbV41Pw4VFGfYhPRcnKjpAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:11:21.904541Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.19618","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3cd8511672405d14ecc171279e9124c786fe5a99a5e92979d33d31327cc6c5ce","sha256:e2e40cc872a6c1dde2beda3c2151aeff57836cc830379e7c1b779c74af83e96a"],"state_sha256":"d8d13e0b61cd2b5c017ba09df4bc71b0f62d2b468de063a6904ff60430ef7789"}