{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RHOUFHQSA4CLCZF452RYRNQMQV","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":"7d1abc49983ac0ea4db77c0927100d89597dba3e7bfe620df96dc46e49fe83a3","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-16T06:25:19Z","title_canon_sha256":"8e671e0939dd929cdc80c159b6f6537dd06efa3f51d1b4c412423a9bce0f89d9"},"schema_version":"1.0","source":{"id":"2607.14614","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.14614","created_at":"2026-07-17T01:21:20Z"},{"alias_kind":"arxiv_version","alias_value":"2607.14614v1","created_at":"2026-07-17T01:21:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14614","created_at":"2026-07-17T01:21:20Z"},{"alias_kind":"pith_short_12","alias_value":"RHOUFHQSA4CL","created_at":"2026-07-17T01:21:20Z"},{"alias_kind":"pith_short_16","alias_value":"RHOUFHQSA4CLCZF4","created_at":"2026-07-17T01:21:20Z"},{"alias_kind":"pith_short_8","alias_value":"RHOUFHQS","created_at":"2026-07-17T01:21:20Z"}],"graph_snapshots":[{"event_id":"sha256:2e799c164a2dfd78c41457d6b3c3f63ae5bc9803138ae2ba63d5b3a5904cfd54","target":"graph","created_at":"2026-07-17T01:21:20Z","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/2607.14614/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) commonly uses entropy for advantage shaping. However, entropy cannot distinguish useful uncertainty from detrimental confusion, limiting its effectiveness as a correctness signal. We propose Contrastive Policy Optimization (CPO), which uses token-level contrastive disagreement between reference-guided and vanilla generation distributions for correctness-aware advantage shaping. Both theoretical and empirical results show that this disagreement reliably indicates token-level correctness. We further show that On-policy Distillation is a speci","authors_text":"Deng Cai, Hao Zhang, Hou Pong Chan, Jia Liu, Long Li, Min Chen, Weiwen Xu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-16T06:25:19Z","title":"Beyond Entropy: Correctness-Aware Advantage Shaping via Contrastive Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14614","kind":"arxiv","version":1},"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":"sha256:7639a52abbb003b25fa376b4707735b8a9b01012b04b724ecb5501436f9c3da5","target":"record","created_at":"2026-07-17T01:21:20Z","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":"7d1abc49983ac0ea4db77c0927100d89597dba3e7bfe620df96dc46e49fe83a3","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-16T06:25:19Z","title_canon_sha256":"8e671e0939dd929cdc80c159b6f6537dd06efa3f51d1b4c412423a9bce0f89d9"},"schema_version":"1.0","source":{"id":"2607.14614","kind":"arxiv","version":1}},"canonical_sha256":"89dd429e120704b164bceea388b60c857ef7436e73fd41d80ca897575abda10b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"89dd429e120704b164bceea388b60c857ef7436e73fd41d80ca897575abda10b","first_computed_at":"2026-07-17T01:21:20.942077Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-17T01:21:20.942077Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZxAzOJuSuTvjNcXdlaCGdGPYn1KcaaCqyeFgK1ewMMWN8VuU9vnfyPGCbHedi4RhgRU4mdgU2q3o3bm2tpLhDA==","signature_status":"signed_v1","signed_at":"2026-07-17T01:21:20.942923Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.14614","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7639a52abbb003b25fa376b4707735b8a9b01012b04b724ecb5501436f9c3da5","sha256:2e799c164a2dfd78c41457d6b3c3f63ae5bc9803138ae2ba63d5b3a5904cfd54"],"state_sha256":"5c99f99249ac22a5c02499ac0b76762a816a65e558c41779e1dedf767b4e51c1"}