{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7KPPWGUDSJEGS575QJLABVGZHM","short_pith_number":"pith:7KPPWGUD","canonical_record":{"source":{"id":"2605.25240","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T19:52:39Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"be3b0afc00c0f5f4790244a0e3b3596507a7ab15d0f048050f3eb52684f1570e","abstract_canon_sha256":"23f44e783321a26705545be1f44f6ab8d4ccbbdc6bb4c0fea051b25d85f0124b"},"schema_version":"1.0"},"canonical_sha256":"fa9efb1a8392486977fd825600d4d93b045ffa6f778771c50bdca9555ee9822c","source":{"kind":"arxiv","id":"2605.25240","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.25240","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"arxiv_version","alias_value":"2605.25240v1","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.25240","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_12","alias_value":"7KPPWGUDSJEG","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_16","alias_value":"7KPPWGUDSJEGS575","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_8","alias_value":"7KPPWGUD","created_at":"2026-05-26T02:04:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7KPPWGUDSJEGS575QJLABVGZHM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.25240","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T19:52:39Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"be3b0afc00c0f5f4790244a0e3b3596507a7ab15d0f048050f3eb52684f1570e","abstract_canon_sha256":"23f44e783321a26705545be1f44f6ab8d4ccbbdc6bb4c0fea051b25d85f0124b"},"schema_version":"1.0"},"canonical_sha256":"fa9efb1a8392486977fd825600d4d93b045ffa6f778771c50bdca9555ee9822c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:24.737194Z","signature_b64":"oblmNL6vOaAZwmilIUzJgOZl79RfP9UioDhbfjf/cHjyd2IjyPmxz6iKog4vQiNQOkQpvh3e/Mgkd7nq1PovDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa9efb1a8392486977fd825600d4d93b045ffa6f778771c50bdca9555ee9822c","last_reissued_at":"2026-05-26T02:04:24.736576Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:24.736576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.25240","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":"2026-05-26T02:04:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MC27B5Ec26/VBeKe1av4ossJ2+DA2lEMaWc34okyuroogmiyALX2viPIWH0nI8RPQ6NoB4ree899PlWpBwQYAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T20:01:04.207389Z"},"content_sha256":"0f2ce79da3a0d94fb643e3846e51fe713a7db71f4ef4cffc8feb6064f37e9fb3","schema_version":"1.0","event_id":"sha256:0f2ce79da3a0d94fb643e3846e51fe713a7db71f4ef4cffc8feb6064f37e9fb3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7KPPWGUDSJEGS575QJLABVGZHM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"JudgmentBench: Comparing Rubric and Preference Evaluation for Quality Assessment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CL","authors_text":"Charles Dickens, Julian Nyarko, Matthew Guillod, Megan Ma, Pierce Kelaita, Riya Ranjan, Ruishi Chen, Russell Yang, Sibo Ma","submitted_at":"2026-05-24T19:52:39Z","abstract_excerpt":"Two methodologies dominate current practices of benchmarking: rubric-based scoring evaluates items against predefined criteria, whereas comparative judgment elicits pairwise preferences between outputs. Although both methodologies are widely used, the choice between them is rarely justified. We release JudgmentBench, a benchmark of 30 real-world legal tasks, paired with 1,539 rubric scores and 1,530 pairwise preference judgments collected from practicing attorneys--including at major U.S. law firms--with substantial experience. The annotations constitute the first publicly available dataset in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.25240","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/2605.25240/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":"2026-05-26T02:04:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IQYtA5FueXgp2eHOGEJvixB5lW+p8Pnxl/P2W6WIqf6lsFvBR5onUglHvTk4jDH1cc7bApIGn10BOFdBUyA5AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T20:01:04.207778Z"},"content_sha256":"033ebb4dfa467c372a86bd9089db654c25602ebbd08c2245cdddfae28950c796","schema_version":"1.0","event_id":"sha256:033ebb4dfa467c372a86bd9089db654c25602ebbd08c2245cdddfae28950c796"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7KPPWGUDSJEGS575QJLABVGZHM/bundle.json","state_url":"https://pith.science/pith/7KPPWGUDSJEGS575QJLABVGZHM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7KPPWGUDSJEGS575QJLABVGZHM/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":"2026-06-01T20:01:04Z","links":{"resolver":"https://pith.science/pith/7KPPWGUDSJEGS575QJLABVGZHM","bundle":"https://pith.science/pith/7KPPWGUDSJEGS575QJLABVGZHM/bundle.json","state":"https://pith.science/pith/7KPPWGUDSJEGS575QJLABVGZHM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7KPPWGUDSJEGS575QJLABVGZHM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7KPPWGUDSJEGS575QJLABVGZHM","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":"23f44e783321a26705545be1f44f6ab8d4ccbbdc6bb4c0fea051b25d85f0124b","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T19:52:39Z","title_canon_sha256":"be3b0afc00c0f5f4790244a0e3b3596507a7ab15d0f048050f3eb52684f1570e"},"schema_version":"1.0","source":{"id":"2605.25240","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.25240","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"arxiv_version","alias_value":"2605.25240v1","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.25240","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_12","alias_value":"7KPPWGUDSJEG","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_16","alias_value":"7KPPWGUDSJEGS575","created_at":"2026-05-26T02:04:24Z"},{"alias_kind":"pith_short_8","alias_value":"7KPPWGUD","created_at":"2026-05-26T02:04:24Z"}],"graph_snapshots":[{"event_id":"sha256:033ebb4dfa467c372a86bd9089db654c25602ebbd08c2245cdddfae28950c796","target":"graph","created_at":"2026-05-26T02:04:24Z","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/2605.25240/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Two methodologies dominate current practices of benchmarking: rubric-based scoring evaluates items against predefined criteria, whereas comparative judgment elicits pairwise preferences between outputs. Although both methodologies are widely used, the choice between them is rarely justified. We release JudgmentBench, a benchmark of 30 real-world legal tasks, paired with 1,539 rubric scores and 1,530 pairwise preference judgments collected from practicing attorneys--including at major U.S. law firms--with substantial experience. The annotations constitute the first publicly available dataset in","authors_text":"Charles Dickens, Julian Nyarko, Matthew Guillod, Megan Ma, Pierce Kelaita, Riya Ranjan, Ruishi Chen, Russell Yang, Sibo Ma","cross_cats":["cs.AI","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T19:52:39Z","title":"JudgmentBench: Comparing Rubric and Preference Evaluation for Quality Assessment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.25240","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":"sha256:0f2ce79da3a0d94fb643e3846e51fe713a7db71f4ef4cffc8feb6064f37e9fb3","target":"record","created_at":"2026-05-26T02:04:24Z","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":"23f44e783321a26705545be1f44f6ab8d4ccbbdc6bb4c0fea051b25d85f0124b","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T19:52:39Z","title_canon_sha256":"be3b0afc00c0f5f4790244a0e3b3596507a7ab15d0f048050f3eb52684f1570e"},"schema_version":"1.0","source":{"id":"2605.25240","kind":"arxiv","version":1}},"canonical_sha256":"fa9efb1a8392486977fd825600d4d93b045ffa6f778771c50bdca9555ee9822c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fa9efb1a8392486977fd825600d4d93b045ffa6f778771c50bdca9555ee9822c","first_computed_at":"2026-05-26T02:04:24.736576Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:04:24.736576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oblmNL6vOaAZwmilIUzJgOZl79RfP9UioDhbfjf/cHjyd2IjyPmxz6iKog4vQiNQOkQpvh3e/Mgkd7nq1PovDg==","signature_status":"signed_v1","signed_at":"2026-05-26T02:04:24.737194Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.25240","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0f2ce79da3a0d94fb643e3846e51fe713a7db71f4ef4cffc8feb6064f37e9fb3","sha256:033ebb4dfa467c372a86bd9089db654c25602ebbd08c2245cdddfae28950c796"],"state_sha256":"703f3eee6be294a957d05d42587a70605893466309dbc9d7f6f9759a4e15ac02"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mwWAImybJoSfXrB0rNghN0Vdh0A5LuEEw3n13oWx2/8Hgc2RYQOJ+JUntU4nujbcLo5IRyqOLPdEK3AlfOSNBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T20:01:04.209992Z","bundle_sha256":"53f08d61f8b47d16959724f776059103c6ad4bd1d95f80cdf0d6a208f0b33054"}}