{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BTMBLFKPJQX4JQDGSTSOQXQW3C","short_pith_number":"pith:BTMBLFKP","schema_version":"1.0","canonical_sha256":"0cd815954f4c2fc4c06694e4e85e16d8a3d88df840cd12b2ef88a13820c3de1a","source":{"kind":"arxiv","id":"2607.00152","version":1},"attestation_state":"computed","paper":{"title":"GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kathleen A. Yearick, Yong Yi Bay","submitted_at":"2026-06-30T20:28:08Z","abstract_excerpt":"Three of the most popular methods for training language models to reason look like three different tricks. They are not. All three adjust a single number: standard deviation, reflecting how much a prompt's sampled answers disagree. When such a model is trained, it answers each problem many times, and an automatic checker marks every answer right or wrong. The standard deviation of those marks measures the disagreement: largest when the answers split evenly between right and wrong, and zero when they all agree. Group Relative Policy Optimization (GRPO) divides by this number, GRPO Done Right (D"},"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":"2607.00152","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-30T20:28:08Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"title_canon_sha256":"968dc1fba3e051dc7b2ab47f5808a7b327b0ea295a4eda96c0828f7da0dbd3a8","abstract_canon_sha256":"d61f1a48e47a4fbadecd334b4cb98ca4974823539c300d94324bbdcebd54ebc0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T00:18:36.803681Z","signature_b64":"31eD27b+YhotPnB/e4jJY/inIFosFAehjcE7lS5bCbPuPPEhZAlTmc+V5fm6xBkm1YiyJA+ureIge+bWRl8PBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0cd815954f4c2fc4c06694e4e85e16d8a3d88df840cd12b2ef88a13820c3de1a","last_reissued_at":"2026-07-02T00:18:36.803099Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T00:18:36.803099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kathleen A. Yearick, Yong Yi Bay","submitted_at":"2026-06-30T20:28:08Z","abstract_excerpt":"Three of the most popular methods for training language models to reason look like three different tricks. They are not. All three adjust a single number: standard deviation, reflecting how much a prompt's sampled answers disagree. When such a model is trained, it answers each problem many times, and an automatic checker marks every answer right or wrong. The standard deviation of those marks measures the disagreement: largest when the answers split evenly between right and wrong, and zero when they all agree. Group Relative Policy Optimization (GRPO) divides by this number, GRPO Done Right (D"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.00152","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/2607.00152/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":"2607.00152","created_at":"2026-07-02T00:18:36.803171+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.00152v1","created_at":"2026-07-02T00:18:36.803171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.00152","created_at":"2026-07-02T00:18:36.803171+00:00"},{"alias_kind":"pith_short_12","alias_value":"BTMBLFKPJQX4","created_at":"2026-07-02T00:18:36.803171+00:00"},{"alias_kind":"pith_short_16","alias_value":"BTMBLFKPJQX4JQDG","created_at":"2026-07-02T00:18:36.803171+00:00"},{"alias_kind":"pith_short_8","alias_value":"BTMBLFKP","created_at":"2026-07-02T00:18:36.803171+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/BTMBLFKPJQX4JQDGSTSOQXQW3C","json":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C.json","graph_json":"https://pith.science/api/pith-number/BTMBLFKPJQX4JQDGSTSOQXQW3C/graph.json","events_json":"https://pith.science/api/pith-number/BTMBLFKPJQX4JQDGSTSOQXQW3C/events.json","paper":"https://pith.science/paper/BTMBLFKP"},"agent_actions":{"view_html":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C","download_json":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C.json","view_paper":"https://pith.science/paper/BTMBLFKP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.00152&json=true","fetch_graph":"https://pith.science/api/pith-number/BTMBLFKPJQX4JQDGSTSOQXQW3C/graph.json","fetch_events":"https://pith.science/api/pith-number/BTMBLFKPJQX4JQDGSTSOQXQW3C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C/action/storage_attestation","attest_author":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C/action/author_attestation","sign_citation":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C/action/citation_signature","submit_replication":"https://pith.science/pith/BTMBLFKPJQX4JQDGSTSOQXQW3C/action/replication_record"}},"created_at":"2026-07-02T00:18:36.803171+00:00","updated_at":"2026-07-02T00:18:36.803171+00:00"}