{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BOJYVLUT3UXFOT3IUR6K3BVFHM","short_pith_number":"pith:BOJYVLUT","canonical_record":{"source":{"id":"2501.10970","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T07:09:11Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"db28c71b89957904bad69fe9910c31de36af51fb91351deda875b86f68c8f2c7","abstract_canon_sha256":"245b81a42ae9753dafbf34725ad5c71378a690667c17a43f6371a7d076f98238"},"schema_version":"1.0"},"canonical_sha256":"0b938aae93dd2e574f68a47cad86a53b1ff179a0f60b0c0402b4db1b5189a36e","source":{"kind":"arxiv","id":"2501.10970","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.10970","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"arxiv_version","alias_value":"2501.10970v4","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10970","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_12","alias_value":"BOJYVLUT3UXF","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_16","alias_value":"BOJYVLUT3UXFOT3I","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_8","alias_value":"BOJYVLUT","created_at":"2026-07-05T11:50:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BOJYVLUT3UXFOT3IUR6K3BVFHM","target":"record","payload":{"canonical_record":{"source":{"id":"2501.10970","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T07:09:11Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"db28c71b89957904bad69fe9910c31de36af51fb91351deda875b86f68c8f2c7","abstract_canon_sha256":"245b81a42ae9753dafbf34725ad5c71378a690667c17a43f6371a7d076f98238"},"schema_version":"1.0"},"canonical_sha256":"0b938aae93dd2e574f68a47cad86a53b1ff179a0f60b0c0402b4db1b5189a36e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:30.109236Z","signature_b64":"1WdKb5PHsHToIhoWpT29Cz+cudW8KN+OU/g8Qnk1EH6mXYI1k92Kg5optHji/RvsKvrhkSIxDmOptZryvsKMDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b938aae93dd2e574f68a47cad86a53b1ff179a0f60b0c0402b4db1b5189a36e","last_reissued_at":"2026-07-05T11:50:30.108754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:30.108754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.10970","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z5y9LR31TR35U2fngFTpEMfiAEFTPmDi1HewwErBIfhpAEcqhpguTJDIryGGXI6Rx6xwVDBCaYTsuCpMG+2CBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:47:50.934403Z"},"content_sha256":"e72a8c73ec8aa96d8451d96a6b539fee3622faab68d7a78e5d63eae5b7bdf35d","schema_version":"1.0","event_id":"sha256:e72a8c73ec8aa96d8451d96a6b539fee3622faab68d7a78e5d63eae5b7bdf35d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BOJYVLUT3UXFOT3IUR6K3BVFHM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMs","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.CL","authors_text":"Nitay Calderon, Roi Reichart, Rotem Dror","submitted_at":"2025-01-19T07:09:11Z","abstract_excerpt":"The \"LLM-as-an-annotator\" and \"LLM-as-a-judge\" paradigms employ Large Language Models (LLMs) as annotators, judges, and evaluators in tasks traditionally performed by humans. LLM annotations are widely used, not only in NLP research but also in fields like medicine, psychology, and social science. Despite their role in shaping study results and insights, there is no standard or rigorous procedure to determine whether LLMs can replace human annotators. In this paper, we propose a novel statistical procedure, the Alternative Annotator Test (alt-test), that requires only a modest subset of annota"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10970","kind":"arxiv","version":4},"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/2501.10970/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":"2026-07-05T11:50:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LDjhCGkgxHPDFBAKH99L/NHlc2GD7mEBaayrc5/+QVLDVwb/uAAJfG9y/UKJ573yQsuqg+7AzL6YwyBRgGB8Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:47:50.934915Z"},"content_sha256":"106b179aae58441cc2ce5b8f43d1e85302a53b2c222961b2766f9fe7220b6f98","schema_version":"1.0","event_id":"sha256:106b179aae58441cc2ce5b8f43d1e85302a53b2c222961b2766f9fe7220b6f98"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/bundle.json","state_url":"https://pith.science/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/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":"2026-08-07T23:47:50Z","links":{"resolver":"https://pith.science/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM","bundle":"https://pith.science/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/bundle.json","state":"https://pith.science/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BOJYVLUT3UXFOT3IUR6K3BVFHM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BOJYVLUT3UXFOT3IUR6K3BVFHM","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":"245b81a42ae9753dafbf34725ad5c71378a690667c17a43f6371a7d076f98238","cross_cats_sorted":["cs.AI","cs.HC"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T07:09:11Z","title_canon_sha256":"db28c71b89957904bad69fe9910c31de36af51fb91351deda875b86f68c8f2c7"},"schema_version":"1.0","source":{"id":"2501.10970","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.10970","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"arxiv_version","alias_value":"2501.10970v4","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10970","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_12","alias_value":"BOJYVLUT3UXF","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_16","alias_value":"BOJYVLUT3UXFOT3I","created_at":"2026-07-05T11:50:30Z"},{"alias_kind":"pith_short_8","alias_value":"BOJYVLUT","created_at":"2026-07-05T11:50:30Z"}],"graph_snapshots":[{"event_id":"sha256:106b179aae58441cc2ce5b8f43d1e85302a53b2c222961b2766f9fe7220b6f98","target":"graph","created_at":"2026-07-05T11:50:30Z","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/2501.10970/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The \"LLM-as-an-annotator\" and \"LLM-as-a-judge\" paradigms employ Large Language Models (LLMs) as annotators, judges, and evaluators in tasks traditionally performed by humans. LLM annotations are widely used, not only in NLP research but also in fields like medicine, psychology, and social science. Despite their role in shaping study results and insights, there is no standard or rigorous procedure to determine whether LLMs can replace human annotators. In this paper, we propose a novel statistical procedure, the Alternative Annotator Test (alt-test), that requires only a modest subset of annota","authors_text":"Nitay Calderon, Roi Reichart, Rotem Dror","cross_cats":["cs.AI","cs.HC"],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T07:09:11Z","title":"The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10970","kind":"arxiv","version":4},"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:e72a8c73ec8aa96d8451d96a6b539fee3622faab68d7a78e5d63eae5b7bdf35d","target":"record","created_at":"2026-07-05T11:50:30Z","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":"245b81a42ae9753dafbf34725ad5c71378a690667c17a43f6371a7d076f98238","cross_cats_sorted":["cs.AI","cs.HC"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T07:09:11Z","title_canon_sha256":"db28c71b89957904bad69fe9910c31de36af51fb91351deda875b86f68c8f2c7"},"schema_version":"1.0","source":{"id":"2501.10970","kind":"arxiv","version":4}},"canonical_sha256":"0b938aae93dd2e574f68a47cad86a53b1ff179a0f60b0c0402b4db1b5189a36e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0b938aae93dd2e574f68a47cad86a53b1ff179a0f60b0c0402b4db1b5189a36e","first_computed_at":"2026-07-05T11:50:30.108754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:50:30.108754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1WdKb5PHsHToIhoWpT29Cz+cudW8KN+OU/g8Qnk1EH6mXYI1k92Kg5optHji/RvsKvrhkSIxDmOptZryvsKMDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:50:30.109236Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.10970","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e72a8c73ec8aa96d8451d96a6b539fee3622faab68d7a78e5d63eae5b7bdf35d","sha256:106b179aae58441cc2ce5b8f43d1e85302a53b2c222961b2766f9fe7220b6f98"],"state_sha256":"052585c3bfc310e3d0cb283d48aba3bb7f2cfa1cb1cbfe56535e874101f31678"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Jl2l33MMqIkOGG3HE3X13aU/Ne7mopg1BUqTOowwz14cS1s5XCDWve9jgBh9IS2EYR2n2rTiadiwRo5MJgP5AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T23:47:50.939880Z","bundle_sha256":"4d481555f7a3a2cff4539db2aa6aecf6ad0ece0d04b808df3192ed2dca9e71a3"}}