{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:M4WYYRWG72XBPX3SWIEDXAJUZJ","short_pith_number":"pith:M4WYYRWG","schema_version":"1.0","canonical_sha256":"672d8c46c6feae17df72b2083b8134ca6d4447295b4285f1d62cde618ae46459","source":{"kind":"arxiv","id":"2607.19313","version":1},"attestation_state":"computed","paper":{"title":"Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aditya Modi, Ankur Samanta, Daniel Jiang, Jalaj Bhandari, Kavosh Asadi, Priyank Agrawal, Shervin Ghasemlou","submitted_at":"2026-07-21T17:28:40Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) improves reasoning in large language models. Yet, typical RLVR approaches fail on difficult problems: when a model cannot generate any correct solutions, it receives \\textit{zero} learning signal. Providing privileged guidance during training, such as solution prefixes, can help overcome this learning cliff by steering the model towards {correct solutions with non-zero reward}. {We call these rollouts \\textit{off-context}: they are generated from a training prompt that contains privileged guidance, while the target objective is defined by t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.19313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-21T17:28:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d7e5352ed458628a7ad3a00bfbbd1a828f42e4c5d787059c6f118c37dce180df","abstract_canon_sha256":"f1a99d5337614eef4dda07a5ff9823e896d576cb43352659b3a8087b8dd313bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-22T01:24:36.199145Z","signature_b64":"Va0T97Q03iQsW0Wj8+N1HuI5Upfv4TKGZvgB1ikInk2UAE3xK1y7UGxTgHzmOlyMsMa4+YOUO0xWTRs6h3eZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"672d8c46c6feae17df72b2083b8134ca6d4447295b4285f1d62cde618ae46459","last_reissued_at":"2026-07-22T01:24:36.198189Z","signature_status":"signed_v1","first_computed_at":"2026-07-22T01:24:36.198189Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aditya Modi, Ankur Samanta, Daniel Jiang, Jalaj Bhandari, Kavosh Asadi, Priyank Agrawal, Shervin Ghasemlou","submitted_at":"2026-07-21T17:28:40Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) improves reasoning in large language models. Yet, typical RLVR approaches fail on difficult problems: when a model cannot generate any correct solutions, it receives \\textit{zero} learning signal. Providing privileged guidance during training, such as solution prefixes, can help overcome this learning cliff by steering the model towards {correct solutions with non-zero reward}. {We call these rollouts \\textit{off-context}: they are generated from a training prompt that contains privileged guidance, while the target objective is defined by t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.19313","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.19313/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.19313","created_at":"2026-07-22T01:24:36.198662+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.19313v1","created_at":"2026-07-22T01:24:36.198662+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.19313","created_at":"2026-07-22T01:24:36.198662+00:00"},{"alias_kind":"pith_short_12","alias_value":"M4WYYRWG72XB","created_at":"2026-07-22T01:24:36.198662+00:00"},{"alias_kind":"pith_short_16","alias_value":"M4WYYRWG72XBPX3S","created_at":"2026-07-22T01:24:36.198662+00:00"},{"alias_kind":"pith_short_8","alias_value":"M4WYYRWG","created_at":"2026-07-22T01:24:36.198662+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ","json":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ.json","graph_json":"https://pith.science/api/pith-number/M4WYYRWG72XBPX3SWIEDXAJUZJ/graph.json","events_json":"https://pith.science/api/pith-number/M4WYYRWG72XBPX3SWIEDXAJUZJ/events.json","paper":"https://pith.science/paper/M4WYYRWG"},"agent_actions":{"view_html":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ","download_json":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ.json","view_paper":"https://pith.science/paper/M4WYYRWG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.19313&json=true","fetch_graph":"https://pith.science/api/pith-number/M4WYYRWG72XBPX3SWIEDXAJUZJ/graph.json","fetch_events":"https://pith.science/api/pith-number/M4WYYRWG72XBPX3SWIEDXAJUZJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ/action/storage_attestation","attest_author":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ/action/author_attestation","sign_citation":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ/action/citation_signature","submit_replication":"https://pith.science/pith/M4WYYRWG72XBPX3SWIEDXAJUZJ/action/replication_record"}},"created_at":"2026-07-22T01:24:36.198662+00:00","updated_at":"2026-07-22T01:24:36.198662+00:00"}