{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:46QXR7ISUIDCCM7QOWZFFVCRMU","short_pith_number":"pith:46QXR7IS","schema_version":"1.0","canonical_sha256":"e7a178fd12a2062133f075b252d451653d1ae476b4b539ee720559c7e2080cbc","source":{"kind":"arxiv","id":"2506.15662","version":1},"attestation_state":"computed","paper":{"title":"CC-LEARN: Cohort-based Consistency Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Avneet Ahuja, Ben Zhou, Jacob Dineen, Ming Shen, Shaswat Shrivastava, Shijie Lu, Xiao Ye, Zhaonan Li, Zhikun Xu","submitted_at":"2025-06-18T17:41:28Z","abstract_excerpt":"Large language models excel at many tasks but still struggle with consistent, robust reasoning. We introduce Cohort-based Consistency Learning (CC-Learn), a reinforcement learning framework that improves the reliability of LLM reasoning by training on cohorts of similar questions derived from shared programmatic abstractions. To enforce cohort-level consistency, we define a composite objective combining cohort accuracy, a retrieval bonus for effective problem decomposition, and a rejection penalty for trivial or invalid lookups that reinforcement learning can directly optimize, unlike supervis"},"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":"2506.15662","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-18T17:41:28Z","cross_cats_sorted":[],"title_canon_sha256":"ebefc7e066ebe33d7d778855ce6d108d3713141e08cfdffbede8eb10d90e6584","abstract_canon_sha256":"55839bc9b727afedae3aea639c88fbc406555f1edf8ecaf9a65cee7ce0f8479b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:48.010023Z","signature_b64":"NByf0o/r6x8RZi3Ncxdt5bh6HCveEku0OJoaOg6U28Fl7rhdhGO1FR73Kx/fdHMH3G2igOAVPeYLZrrTGxbODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7a178fd12a2062133f075b252d451653d1ae476b4b539ee720559c7e2080cbc","last_reissued_at":"2026-07-05T11:23:48.009519Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:48.009519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CC-LEARN: Cohort-based Consistency Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Avneet Ahuja, Ben Zhou, Jacob Dineen, Ming Shen, Shaswat Shrivastava, Shijie Lu, Xiao Ye, Zhaonan Li, Zhikun Xu","submitted_at":"2025-06-18T17:41:28Z","abstract_excerpt":"Large language models excel at many tasks but still struggle with consistent, robust reasoning. We introduce Cohort-based Consistency Learning (CC-Learn), a reinforcement learning framework that improves the reliability of LLM reasoning by training on cohorts of similar questions derived from shared programmatic abstractions. To enforce cohort-level consistency, we define a composite objective combining cohort accuracy, a retrieval bonus for effective problem decomposition, and a rejection penalty for trivial or invalid lookups that reinforcement learning can directly optimize, unlike supervis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15662","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/2506.15662/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":"2506.15662","created_at":"2026-07-05T11:23:48.009583+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.15662v1","created_at":"2026-07-05T11:23:48.009583+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15662","created_at":"2026-07-05T11:23:48.009583+00:00"},{"alias_kind":"pith_short_12","alias_value":"46QXR7ISUIDC","created_at":"2026-07-05T11:23:48.009583+00:00"},{"alias_kind":"pith_short_16","alias_value":"46QXR7ISUIDCCM7Q","created_at":"2026-07-05T11:23:48.009583+00:00"},{"alias_kind":"pith_short_8","alias_value":"46QXR7IS","created_at":"2026-07-05T11:23:48.009583+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.10746","citing_title":"RECAP: Transparent Inference-Time Emotion Alignment for Medical Dialogue Systems","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU","json":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU.json","graph_json":"https://pith.science/api/pith-number/46QXR7ISUIDCCM7QOWZFFVCRMU/graph.json","events_json":"https://pith.science/api/pith-number/46QXR7ISUIDCCM7QOWZFFVCRMU/events.json","paper":"https://pith.science/paper/46QXR7IS"},"agent_actions":{"view_html":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU","download_json":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU.json","view_paper":"https://pith.science/paper/46QXR7IS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.15662&json=true","fetch_graph":"https://pith.science/api/pith-number/46QXR7ISUIDCCM7QOWZFFVCRMU/graph.json","fetch_events":"https://pith.science/api/pith-number/46QXR7ISUIDCCM7QOWZFFVCRMU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU/action/storage_attestation","attest_author":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU/action/author_attestation","sign_citation":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU/action/citation_signature","submit_replication":"https://pith.science/pith/46QXR7ISUIDCCM7QOWZFFVCRMU/action/replication_record"}},"created_at":"2026-07-05T11:23:48.009583+00:00","updated_at":"2026-07-05T11:23:48.009583+00:00"}