{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Q4XX4VTJE3DBEOOCKLESVTCGVA","short_pith_number":"pith:Q4XX4VTJ","canonical_record":{"source":{"id":"2605.27483","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T13:02:35Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"191880a6a018bb000b94dab828bcf52f1a741768800a430375bc4ce6bba66f49","abstract_canon_sha256":"9c0fff979b164830d0ce9e1e78974316497d6acf8958642659f9acb8fe86321d"},"schema_version":"1.0"},"canonical_sha256":"872f7e566926c61239c252c92acc46a827977fa39297975c005405d32ca12b3e","source":{"kind":"arxiv","id":"2605.27483","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27483","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27483v1","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27483","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"Q4XX4VTJE3DB","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"Q4XX4VTJE3DBEOOC","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"Q4XX4VTJ","created_at":"2026-05-28T00:05:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Q4XX4VTJE3DBEOOCKLESVTCGVA","target":"record","payload":{"canonical_record":{"source":{"id":"2605.27483","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T13:02:35Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"191880a6a018bb000b94dab828bcf52f1a741768800a430375bc4ce6bba66f49","abstract_canon_sha256":"9c0fff979b164830d0ce9e1e78974316497d6acf8958642659f9acb8fe86321d"},"schema_version":"1.0"},"canonical_sha256":"872f7e566926c61239c252c92acc46a827977fa39297975c005405d32ca12b3e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T00:05:20.647820Z","signature_b64":"Yo03X8r1fdEAUPyrmA3aX9e9cQjILoQKqmz3Gxuos4VcUXQNpu8TGjqpXix11WQ/l/zO7Nz2sm1I8TMPbVyeBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"872f7e566926c61239c252c92acc46a827977fa39297975c005405d32ca12b3e","last_reissued_at":"2026-05-28T00:05:20.647113Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T00:05:20.647113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.27483","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":"2026-05-28T00:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Vuz0PHe5TcCTqEWp26FfSvPkUaHC9O1a14lFo69TxKOf4CxyicDXaWhlsq78m2XTxoeMa/BGjkmIjj1ZrYNeBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T04:57:11.682811Z"},"content_sha256":"d4cc3096a6516a02efb5cda41c3162c27d9ab71472550171049c96130062f841","schema_version":"1.0","event_id":"sha256:d4cc3096a6516a02efb5cda41c3162c27d9ab71472550171049c96130062f841"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Q4XX4VTJE3DBEOOCKLESVTCGVA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Debate Helps Weak Judges Reward Stronger Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ethan Elasky, Frank Nakasako, Naman Goyal","submitted_at":"2026-05-26T13:02:35Z","abstract_excerpt":"Despite theoretical promise, debate as a scalable oversight protocol has produced mixed empirical results: gains in some settings, and null effects in others, especially when the judge does not have information hidden from it. We study proposer-critic debate in a stronger-debater/weaker-judge setting on programmatically verifiable code and logic tasks. Debate helps the judge over a consultancy baseline when the critic provides a usable advantage: the critic's classification ability must exceed the judge's, and the judge must treat critic speeches as claims to verify rather than testimony to su"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27483","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/2605.27483/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":"2026-05-28T00:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V/c3WdHs0v9Wd0d9Uj0fqdRCarzKCXGazGzJYUmn6Oi1BuoRs8M1eNoBebq0z9o3QULiTsRrr+8+lLIGlwdxBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T04:57:11.683249Z"},"content_sha256":"1687c2eb7e8c10bd3b4e6c4dbecc7c20ad9071b9e7aa758803ccbc25a8c317c1","schema_version":"1.0","event_id":"sha256:1687c2eb7e8c10bd3b4e6c4dbecc7c20ad9071b9e7aa758803ccbc25a8c317c1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/bundle.json","state_url":"https://pith.science/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/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":"2026-06-07T04:57:11Z","links":{"resolver":"https://pith.science/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA","bundle":"https://pith.science/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/bundle.json","state":"https://pith.science/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q4XX4VTJE3DBEOOCKLESVTCGVA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Q4XX4VTJE3DBEOOCKLESVTCGVA","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":"9c0fff979b164830d0ce9e1e78974316497d6acf8958642659f9acb8fe86321d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T13:02:35Z","title_canon_sha256":"191880a6a018bb000b94dab828bcf52f1a741768800a430375bc4ce6bba66f49"},"schema_version":"1.0","source":{"id":"2605.27483","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27483","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27483v1","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27483","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"Q4XX4VTJE3DB","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"Q4XX4VTJE3DBEOOC","created_at":"2026-05-28T00:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"Q4XX4VTJ","created_at":"2026-05-28T00:05:20Z"}],"graph_snapshots":[{"event_id":"sha256:1687c2eb7e8c10bd3b4e6c4dbecc7c20ad9071b9e7aa758803ccbc25a8c317c1","target":"graph","created_at":"2026-05-28T00:05:20Z","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/2605.27483/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite theoretical promise, debate as a scalable oversight protocol has produced mixed empirical results: gains in some settings, and null effects in others, especially when the judge does not have information hidden from it. We study proposer-critic debate in a stronger-debater/weaker-judge setting on programmatically verifiable code and logic tasks. Debate helps the judge over a consultancy baseline when the critic provides a usable advantage: the critic's classification ability must exceed the judge's, and the judge must treat critic speeches as claims to verify rather than testimony to su","authors_text":"Ethan Elasky, Frank Nakasako, Naman Goyal","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T13:02:35Z","title":"Debate Helps Weak Judges Reward Stronger Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27483","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:d4cc3096a6516a02efb5cda41c3162c27d9ab71472550171049c96130062f841","target":"record","created_at":"2026-05-28T00:05:20Z","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":"9c0fff979b164830d0ce9e1e78974316497d6acf8958642659f9acb8fe86321d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T13:02:35Z","title_canon_sha256":"191880a6a018bb000b94dab828bcf52f1a741768800a430375bc4ce6bba66f49"},"schema_version":"1.0","source":{"id":"2605.27483","kind":"arxiv","version":1}},"canonical_sha256":"872f7e566926c61239c252c92acc46a827977fa39297975c005405d32ca12b3e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"872f7e566926c61239c252c92acc46a827977fa39297975c005405d32ca12b3e","first_computed_at":"2026-05-28T00:05:20.647113Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T00:05:20.647113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Yo03X8r1fdEAUPyrmA3aX9e9cQjILoQKqmz3Gxuos4VcUXQNpu8TGjqpXix11WQ/l/zO7Nz2sm1I8TMPbVyeBA==","signature_status":"signed_v1","signed_at":"2026-05-28T00:05:20.647820Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.27483","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4cc3096a6516a02efb5cda41c3162c27d9ab71472550171049c96130062f841","sha256:1687c2eb7e8c10bd3b4e6c4dbecc7c20ad9071b9e7aa758803ccbc25a8c317c1"],"state_sha256":"d89fec4389a6811f8dde0ad5334ad8a9af00f7a6684f189df4f7743b868d5c30"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yw7lVv2ZFvrgc8JjfN1IMwEybiLdKWM3C3oYFLOp70BfxLD7incMkBoUskwr8Pm5+HBYAjr97LX+sS/OolKDAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T04:57:11.686050Z","bundle_sha256":"2858dfdb80d4b2176cd94df09a2bceed9dc99147c08ce06883529488927e592b"}}