{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ASQSBGW2TJNHKVXQJQRRMQ6LVY","short_pith_number":"pith:ASQSBGW2","schema_version":"1.0","canonical_sha256":"04a1209ada9a5a7556f04c231643cbae3b6ddef59679fae79dd39ce25a6f675a","source":{"kind":"arxiv","id":"2607.16850","version":1},"attestation_state":"computed","paper":{"title":"Group Entropy-Controlled Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengqi Lyu, Guangran Cheng, Kai Chen, Songyang Gao, Wenwei Zhang","submitted_at":"2026-07-18T15:13:33Z","abstract_excerpt":"Entropy control has become an effective tool in reinforcement learning (RL) of large language models (LLMs), helping balance exploration-exploitation trade-off during alignment process. Such RL paradigm is often conducted on mixtures of heterogeneous tasks, which induce distinct entropy regimes under the same policy, making global or token-level entropy regulation insufficient to corresponding heterogeneous needs of exploration. This heterogeneity further makes GRPO-style normalized advantages induce an entropy-dependent bias, making advantage signals across prompt groups statistically non-com"},"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":"2607.16850","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-18T15:13:33Z","cross_cats_sorted":[],"title_canon_sha256":"013343d05a7bdc4e2786de0552375c289ef68b9259179e85b827cdf95b61797e","abstract_canon_sha256":"31e0b23eba8b75456999f0f8ffc3864710bb85299b7de49e6f054e8f728eff23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:02.179850Z","signature_b64":"hGDvyWA3941VOb2lWARGDOqUl+tvyzxsCTotP89IdalL8GpMx7QCvOdsc0bSS9GO+YXn6jqt5es5hJmkZIXsBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04a1209ada9a5a7556f04c231643cbae3b6ddef59679fae79dd39ce25a6f675a","last_reissued_at":"2026-07-21T01:21:02.179004Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:02.179004Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Group Entropy-Controlled Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengqi Lyu, Guangran Cheng, Kai Chen, Songyang Gao, Wenwei Zhang","submitted_at":"2026-07-18T15:13:33Z","abstract_excerpt":"Entropy control has become an effective tool in reinforcement learning (RL) of large language models (LLMs), helping balance exploration-exploitation trade-off during alignment process. Such RL paradigm is often conducted on mixtures of heterogeneous tasks, which induce distinct entropy regimes under the same policy, making global or token-level entropy regulation insufficient to corresponding heterogeneous needs of exploration. This heterogeneity further makes GRPO-style normalized advantages induce an entropy-dependent bias, making advantage signals across prompt groups statistically non-com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.16850","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/2607.16850/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":"2607.16850","created_at":"2026-07-21T01:21:02.179439+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.16850v1","created_at":"2026-07-21T01:21:02.179439+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.16850","created_at":"2026-07-21T01:21:02.179439+00:00"},{"alias_kind":"pith_short_12","alias_value":"ASQSBGW2TJNH","created_at":"2026-07-21T01:21:02.179439+00:00"},{"alias_kind":"pith_short_16","alias_value":"ASQSBGW2TJNHKVXQ","created_at":"2026-07-21T01:21:02.179439+00:00"},{"alias_kind":"pith_short_8","alias_value":"ASQSBGW2","created_at":"2026-07-21T01:21:02.179439+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/ASQSBGW2TJNHKVXQJQRRMQ6LVY","json":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY.json","graph_json":"https://pith.science/api/pith-number/ASQSBGW2TJNHKVXQJQRRMQ6LVY/graph.json","events_json":"https://pith.science/api/pith-number/ASQSBGW2TJNHKVXQJQRRMQ6LVY/events.json","paper":"https://pith.science/paper/ASQSBGW2"},"agent_actions":{"view_html":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY","download_json":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY.json","view_paper":"https://pith.science/paper/ASQSBGW2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.16850&json=true","fetch_graph":"https://pith.science/api/pith-number/ASQSBGW2TJNHKVXQJQRRMQ6LVY/graph.json","fetch_events":"https://pith.science/api/pith-number/ASQSBGW2TJNHKVXQJQRRMQ6LVY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY/action/storage_attestation","attest_author":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY/action/author_attestation","sign_citation":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY/action/citation_signature","submit_replication":"https://pith.science/pith/ASQSBGW2TJNHKVXQJQRRMQ6LVY/action/replication_record"}},"created_at":"2026-07-21T01:21:02.179439+00:00","updated_at":"2026-07-21T01:21:02.179439+00:00"}