{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PJ3CBNLBUDJGEWQZZ2WVUWCQYR","short_pith_number":"pith:PJ3CBNLB","canonical_record":{"source":{"id":"2508.05613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:53:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"58f61a94f79a3a59141d18f832de5f5c50f486dd4c2c912570d1b4cc0dd0c6c2","abstract_canon_sha256":"91a5ba981464e713a5af8a12631902a3f44aef8f71e40813a1f6ddf9e1388152"},"schema_version":"1.0"},"canonical_sha256":"7a7620b561a0d2625a19cead5a5850c44686659943902dc7ab7cd273dd15b534","source":{"kind":"arxiv","id":"2508.05613","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.05613","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"arxiv_version","alias_value":"2508.05613v1","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05613","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_12","alias_value":"PJ3CBNLBUDJG","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_16","alias_value":"PJ3CBNLBUDJGEWQZ","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_8","alias_value":"PJ3CBNLB","created_at":"2026-07-05T11:50:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PJ3CBNLBUDJGEWQZZ2WVUWCQYR","target":"record","payload":{"canonical_record":{"source":{"id":"2508.05613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:53:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"58f61a94f79a3a59141d18f832de5f5c50f486dd4c2c912570d1b4cc0dd0c6c2","abstract_canon_sha256":"91a5ba981464e713a5af8a12631902a3f44aef8f71e40813a1f6ddf9e1388152"},"schema_version":"1.0"},"canonical_sha256":"7a7620b561a0d2625a19cead5a5850c44686659943902dc7ab7cd273dd15b534","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:22.485936Z","signature_b64":"+rOSLq5xoePUr46icC9uyQugKgiNRX7Mz5QCt2h2jKc3Se6t7UWGKVFwCLidzQNTmgaDX9MonR0eikhjTS/tBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a7620b561a0d2625a19cead5a5850c44686659943902dc7ab7cd273dd15b534","last_reissued_at":"2026-07-05T11:50:22.485420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:22.485420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.05613","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"103d0ccml6BmBr26JQTsemAxahFCwSEUTdbGQXqmJiLNyjgfC7orbFDPsvGKRN0EWeRP1AscbLK5WapbgAlUCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T11:09:11.559865Z"},"content_sha256":"6b6b35042b5aa97d55e1d555c435718177681916ddb613f3418ff7904626e3bd","schema_version":"1.0","event_id":"sha256:6b6b35042b5aa97d55e1d555c435718177681916ddb613f3418ff7904626e3bd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PJ3CBNLBUDJGEWQZZ2WVUWCQYR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guiyang Hou, Haitao Hong, Jun Xiao, Weiming Lu, Wenqi Zhang, Xingyu Wu, Yongliang Shen, Yuchen Yan","submitted_at":"2025-08-07T17:53:56Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performance in reasoning tasks, where reinforcement learning (RL) serves as a key algorithm for enhancing their reasoning capabilities. Currently, there are two mainstream reward paradigms: model-based rewards and rule-based rewards. However, both approaches suffer from limitations: rule-based rewards lack robustness, while model-based rewards are vulnerable to reward hacking. To address these issues, we propose Cooper(Co-optimizing Policy Model and Reward Model), a RL framework that jointly optimizes both the policy model and the rewar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05613","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/2508.05613/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yfnYAUDSywxPH5UG0n+vIarBBdGZRVcbqkJNyhgMhh0heD0g8WC/Z3vi4zqWDudqlIv09QBLXTl94fDIKFKcAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T11:09:11.560227Z"},"content_sha256":"62c297f5e5b12868b7dcf785e0bd03796f056ec263264b17b9346d8417c069a6","schema_version":"1.0","event_id":"sha256:62c297f5e5b12868b7dcf785e0bd03796f056ec263264b17b9346d8417c069a6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/bundle.json","state_url":"https://pith.science/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T11:09:11Z","links":{"resolver":"https://pith.science/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR","bundle":"https://pith.science/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/bundle.json","state":"https://pith.science/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PJ3CBNLBUDJGEWQZZ2WVUWCQYR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PJ3CBNLBUDJGEWQZZ2WVUWCQYR","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":"91a5ba981464e713a5af8a12631902a3f44aef8f71e40813a1f6ddf9e1388152","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:53:56Z","title_canon_sha256":"58f61a94f79a3a59141d18f832de5f5c50f486dd4c2c912570d1b4cc0dd0c6c2"},"schema_version":"1.0","source":{"id":"2508.05613","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.05613","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"arxiv_version","alias_value":"2508.05613v1","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05613","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_12","alias_value":"PJ3CBNLBUDJG","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_16","alias_value":"PJ3CBNLBUDJGEWQZ","created_at":"2026-07-05T11:50:22Z"},{"alias_kind":"pith_short_8","alias_value":"PJ3CBNLB","created_at":"2026-07-05T11:50:22Z"}],"graph_snapshots":[{"event_id":"sha256:62c297f5e5b12868b7dcf785e0bd03796f056ec263264b17b9346d8417c069a6","target":"graph","created_at":"2026-07-05T11:50:22Z","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/2508.05613/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performance in reasoning tasks, where reinforcement learning (RL) serves as a key algorithm for enhancing their reasoning capabilities. Currently, there are two mainstream reward paradigms: model-based rewards and rule-based rewards. However, both approaches suffer from limitations: rule-based rewards lack robustness, while model-based rewards are vulnerable to reward hacking. To address these issues, we propose Cooper(Co-optimizing Policy Model and Reward Model), a RL framework that jointly optimizes both the policy model and the rewar","authors_text":"Guiyang Hou, Haitao Hong, Jun Xiao, Weiming Lu, Wenqi Zhang, Xingyu Wu, Yongliang Shen, Yuchen Yan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:53:56Z","title":"Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05613","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":"sha256:6b6b35042b5aa97d55e1d555c435718177681916ddb613f3418ff7904626e3bd","target":"record","created_at":"2026-07-05T11:50:22Z","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":"91a5ba981464e713a5af8a12631902a3f44aef8f71e40813a1f6ddf9e1388152","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T17:53:56Z","title_canon_sha256":"58f61a94f79a3a59141d18f832de5f5c50f486dd4c2c912570d1b4cc0dd0c6c2"},"schema_version":"1.0","source":{"id":"2508.05613","kind":"arxiv","version":1}},"canonical_sha256":"7a7620b561a0d2625a19cead5a5850c44686659943902dc7ab7cd273dd15b534","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a7620b561a0d2625a19cead5a5850c44686659943902dc7ab7cd273dd15b534","first_computed_at":"2026-07-05T11:50:22.485420Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:50:22.485420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+rOSLq5xoePUr46icC9uyQugKgiNRX7Mz5QCt2h2jKc3Se6t7UWGKVFwCLidzQNTmgaDX9MonR0eikhjTS/tBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:50:22.485936Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.05613","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6b6b35042b5aa97d55e1d555c435718177681916ddb613f3418ff7904626e3bd","sha256:62c297f5e5b12868b7dcf785e0bd03796f056ec263264b17b9346d8417c069a6"],"state_sha256":"5bb0f75b078fdaad07e12c1436894ffe987a68332894e9ec9e57b3f22ace280b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OyjOl+tz1uO7DHFymFBe32aMTzRVm0Tvcill9IYqEhluKsIrfFjeqgNQgcraz/aE+W0iepyy1ZLeOslKsyz/AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T11:09:11.562180Z","bundle_sha256":"4d1f65fdea241c099aacfb29e6adbe95155c49fb84ed26bbb879aa4dff5c2253"}}