{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4YJNBQ7GN3WB4ZSFMWFLQRKYQ3","short_pith_number":"pith:4YJNBQ7G","canonical_record":{"source":{"id":"2507.09104","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-12T01:34:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7ebb1f52505e838a6f2a2274198cd5fbb8b3db1e68c1d03b54daa124981450b8","abstract_canon_sha256":"24a65b9b9af3f353ff82567ed35a3c63b7a330d6cc55a90c6d283681e326a75f"},"schema_version":"1.0"},"canonical_sha256":"e612d0c3e66eec1e6645658ab8455886ca98dd0929960e395769cecf4b3ea179","source":{"kind":"arxiv","id":"2507.09104","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.09104","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"arxiv_version","alias_value":"2507.09104v1","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.09104","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_12","alias_value":"4YJNBQ7GN3WB","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_16","alias_value":"4YJNBQ7GN3WB4ZSF","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_8","alias_value":"4YJNBQ7G","created_at":"2026-07-05T11:36:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4YJNBQ7GN3WB4ZSFMWFLQRKYQ3","target":"record","payload":{"canonical_record":{"source":{"id":"2507.09104","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-12T01:34:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7ebb1f52505e838a6f2a2274198cd5fbb8b3db1e68c1d03b54daa124981450b8","abstract_canon_sha256":"24a65b9b9af3f353ff82567ed35a3c63b7a330d6cc55a90c6d283681e326a75f"},"schema_version":"1.0"},"canonical_sha256":"e612d0c3e66eec1e6645658ab8455886ca98dd0929960e395769cecf4b3ea179","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:36:03.740422Z","signature_b64":"KlS3t5r8UOPzDPGMvEQBDvUPWgdbTkICfatiSR6KVpFyTjo2IzDX+I5tsrALdO+/jxppeOOBWQZpteOHDCHBCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e612d0c3e66eec1e6645658ab8455886ca98dd0929960e395769cecf4b3ea179","last_reissued_at":"2026-07-05T11:36:03.739887Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:36:03.739887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.09104","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":"2026-07-05T11:36:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5mDig5oT9SXR320w61yRXhawaPqTqV42Tf0el6wtUHGF2LukCnrUL0yK98y6CKSJvL80Ea6EFezxSHgAq+H5AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:12:28.628030Z"},"content_sha256":"11c88cd2f50e6767a245c7d2fddcbd87810d73628bbec79b14549529d473cb8a","schema_version":"1.0","event_id":"sha256:11c88cd2f50e6767a245c7d2fddcbd87810d73628bbec79b14549529d473cb8a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4YJNBQ7GN3WB4ZSFMWFLQRKYQ3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CompassJudger-2: Towards Generalist Judge Model via Verifiable Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexander Lam, Kai Chen, Maosong Cao, Songyang Zhang, Taolin Zhang","submitted_at":"2025-07-12T01:34:24Z","abstract_excerpt":"Recently, the role of LLM-as-judge in evaluating large language models has gained prominence. However, current judge models suffer from narrow specialization and limited robustness, undermining their capacity for comprehensive evaluations. In this work, we present CompassJudger-2, a novel generalist judge model that overcomes these limitations via a task-driven, multi-domain data curation strategy. Central to our approach is supervising judgment tasks with verifiable rewards, guiding intrinsic critical reasoning through rejection sampling to foster robust, generalizable judgment capabilities. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.09104","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/2507.09104/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":"2026-07-05T11:36:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GVnYnCBctR/EVBcvlacz+x9UJINZMODHYwYozzHrVJh09DmatvxsY8su22+C7w8vZtv18Rv9In6mQiIKDid6CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:12:28.628583Z"},"content_sha256":"347e22919a045631e0cc04c4ebcde9be2f07abcbd88e3c3f028229f176137f72","schema_version":"1.0","event_id":"sha256:347e22919a045631e0cc04c4ebcde9be2f07abcbd88e3c3f028229f176137f72"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/bundle.json","state_url":"https://pith.science/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/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":"2026-08-10T03:12:28Z","links":{"resolver":"https://pith.science/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3","bundle":"https://pith.science/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/bundle.json","state":"https://pith.science/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4YJNBQ7GN3WB4ZSFMWFLQRKYQ3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4YJNBQ7GN3WB4ZSFMWFLQRKYQ3","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":"24a65b9b9af3f353ff82567ed35a3c63b7a330d6cc55a90c6d283681e326a75f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-12T01:34:24Z","title_canon_sha256":"7ebb1f52505e838a6f2a2274198cd5fbb8b3db1e68c1d03b54daa124981450b8"},"schema_version":"1.0","source":{"id":"2507.09104","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.09104","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"arxiv_version","alias_value":"2507.09104v1","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.09104","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_12","alias_value":"4YJNBQ7GN3WB","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_16","alias_value":"4YJNBQ7GN3WB4ZSF","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_8","alias_value":"4YJNBQ7G","created_at":"2026-07-05T11:36:03Z"}],"graph_snapshots":[{"event_id":"sha256:347e22919a045631e0cc04c4ebcde9be2f07abcbd88e3c3f028229f176137f72","target":"graph","created_at":"2026-07-05T11:36:03Z","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/2507.09104/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recently, the role of LLM-as-judge in evaluating large language models has gained prominence. However, current judge models suffer from narrow specialization and limited robustness, undermining their capacity for comprehensive evaluations. In this work, we present CompassJudger-2, a novel generalist judge model that overcomes these limitations via a task-driven, multi-domain data curation strategy. Central to our approach is supervising judgment tasks with verifiable rewards, guiding intrinsic critical reasoning through rejection sampling to foster robust, generalizable judgment capabilities. ","authors_text":"Alexander Lam, Kai Chen, Maosong Cao, Songyang Zhang, Taolin Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-12T01:34:24Z","title":"CompassJudger-2: Towards Generalist Judge Model via Verifiable Rewards"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.09104","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":"sha256:11c88cd2f50e6767a245c7d2fddcbd87810d73628bbec79b14549529d473cb8a","target":"record","created_at":"2026-07-05T11:36:03Z","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":"24a65b9b9af3f353ff82567ed35a3c63b7a330d6cc55a90c6d283681e326a75f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-12T01:34:24Z","title_canon_sha256":"7ebb1f52505e838a6f2a2274198cd5fbb8b3db1e68c1d03b54daa124981450b8"},"schema_version":"1.0","source":{"id":"2507.09104","kind":"arxiv","version":1}},"canonical_sha256":"e612d0c3e66eec1e6645658ab8455886ca98dd0929960e395769cecf4b3ea179","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e612d0c3e66eec1e6645658ab8455886ca98dd0929960e395769cecf4b3ea179","first_computed_at":"2026-07-05T11:36:03.739887Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:36:03.739887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KlS3t5r8UOPzDPGMvEQBDvUPWgdbTkICfatiSR6KVpFyTjo2IzDX+I5tsrALdO+/jxppeOOBWQZpteOHDCHBCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:36:03.740422Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.09104","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:11c88cd2f50e6767a245c7d2fddcbd87810d73628bbec79b14549529d473cb8a","sha256:347e22919a045631e0cc04c4ebcde9be2f07abcbd88e3c3f028229f176137f72"],"state_sha256":"4efe59183cf79518faf895c9b6f9a66df8b7a98248d16f6f6281e2bf8f5337df"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4hxv5Gkv7GTlmplyoJFaroAYAMnkMtFjteeT1JsfAOiBKB4xFiHRzr81VNKHryWcSWup9UkBYimL2UC7S70zDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T03:12:28.634136Z","bundle_sha256":"b5c8c44f6c3b2d680e59a6eebf8ab41b6404b96484aacec36bb74bd8d7a8bff9"}}