{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QJFVZJSL3MAFJLAP2A5PCRXF4Z","short_pith_number":"pith:QJFVZJSL","canonical_record":{"source":{"id":"2411.00640","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2024-11-01T14:57:16Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"986f5ba6fd4cc287cda3aa767e08afe4355d452bf75f797bbb8058e0576b8611","abstract_canon_sha256":"dd4eda217022718702bbebbbf50a92212361e54fbfa9795a508374e4f4f31939"},"schema_version":"1.0"},"canonical_sha256":"824b5ca64bdb0054ac0fd03af146e5e645f3df353043df0bcd1903421b1d66fd","source":{"kind":"arxiv","id":"2411.00640","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.00640","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"arxiv_version","alias_value":"2411.00640v1","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.00640","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_12","alias_value":"QJFVZJSL3MAF","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_16","alias_value":"QJFVZJSL3MAFJLAP","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_8","alias_value":"QJFVZJSL","created_at":"2026-07-05T09:29:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QJFVZJSL3MAFJLAP2A5PCRXF4Z","target":"record","payload":{"canonical_record":{"source":{"id":"2411.00640","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2024-11-01T14:57:16Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"986f5ba6fd4cc287cda3aa767e08afe4355d452bf75f797bbb8058e0576b8611","abstract_canon_sha256":"dd4eda217022718702bbebbbf50a92212361e54fbfa9795a508374e4f4f31939"},"schema_version":"1.0"},"canonical_sha256":"824b5ca64bdb0054ac0fd03af146e5e645f3df353043df0bcd1903421b1d66fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:46.059457Z","signature_b64":"dAnAScsWYs0wJ/OhNl1XRAOCghbCN387SvqprAO2zzN28P7RCbs5zvlTHaG63iLAwpKTr/FgYcqNTfGR8742Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"824b5ca64bdb0054ac0fd03af146e5e645f3df353043df0bcd1903421b1d66fd","last_reissued_at":"2026-07-05T09:29:46.058980Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:46.058980Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.00640","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":"2026-07-05T09:29:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TUOM0cPBt1peurHW6RqfL0t5wG8CPbHZ1Nlg17ov047v6i8aGp95K1tuTcZKvr4CU7ZLN59Tp93E7F9bmaLzBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T01:41:58.223339Z"},"content_sha256":"dce7700d5acd56878ecf1803e575389ed8db7ec4e6af81a09c21fbfc45af056e","schema_version":"1.0","event_id":"sha256:dce7700d5acd56878ecf1803e575389ed8db7ec4e6af81a09c21fbfc45af056e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QJFVZJSL3MAFJLAP2A5PCRXF4Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"stat.AP","authors_text":"Evan Miller","submitted_at":"2024-11-01T14:57:16Z","abstract_excerpt":"Evaluations are critical for understanding the capabilities of large language models (LLMs). Fundamentally, evaluations are experiments; but the literature on evaluations has largely ignored the literature from other sciences on experiment analysis and planning. This article shows researchers with some training in statistics how to think about and analyze data from language model evaluations. Conceptualizing evaluation questions as having been drawn from an unseen super-population, we present formulas for analyzing evaluation data, measuring differences between two models, and planning an eval"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.00640","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/2411.00640/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":"2026-07-05T09:29:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xdHOrOVi0vN2psBu+gYgZPK//FdL2dhJHqeH9o4jQvy6nz86MAkB4WSzOWacnp1OgRerKi3dPDJ+IIk2eWDmDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T01:41:58.223748Z"},"content_sha256":"76750dfc15db7905afa606b3f0e912fc7794f7e725a364dfaaa1ee9bde06bea7","schema_version":"1.0","event_id":"sha256:76750dfc15db7905afa606b3f0e912fc7794f7e725a364dfaaa1ee9bde06bea7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/bundle.json","state_url":"https://pith.science/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/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":"2026-07-10T01:41:58Z","links":{"resolver":"https://pith.science/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z","bundle":"https://pith.science/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/bundle.json","state":"https://pith.science/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QJFVZJSL3MAFJLAP2A5PCRXF4Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QJFVZJSL3MAFJLAP2A5PCRXF4Z","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":"dd4eda217022718702bbebbbf50a92212361e54fbfa9795a508374e4f4f31939","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2024-11-01T14:57:16Z","title_canon_sha256":"986f5ba6fd4cc287cda3aa767e08afe4355d452bf75f797bbb8058e0576b8611"},"schema_version":"1.0","source":{"id":"2411.00640","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.00640","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"arxiv_version","alias_value":"2411.00640v1","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.00640","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_12","alias_value":"QJFVZJSL3MAF","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_16","alias_value":"QJFVZJSL3MAFJLAP","created_at":"2026-07-05T09:29:46Z"},{"alias_kind":"pith_short_8","alias_value":"QJFVZJSL","created_at":"2026-07-05T09:29:46Z"}],"graph_snapshots":[{"event_id":"sha256:76750dfc15db7905afa606b3f0e912fc7794f7e725a364dfaaa1ee9bde06bea7","target":"graph","created_at":"2026-07-05T09:29:46Z","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/2411.00640/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluations are critical for understanding the capabilities of large language models (LLMs). Fundamentally, evaluations are experiments; but the literature on evaluations has largely ignored the literature from other sciences on experiment analysis and planning. This article shows researchers with some training in statistics how to think about and analyze data from language model evaluations. Conceptualizing evaluation questions as having been drawn from an unseen super-population, we present formulas for analyzing evaluation data, measuring differences between two models, and planning an eval","authors_text":"Evan Miller","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2024-11-01T14:57:16Z","title":"Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.00640","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:dce7700d5acd56878ecf1803e575389ed8db7ec4e6af81a09c21fbfc45af056e","target":"record","created_at":"2026-07-05T09:29:46Z","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":"dd4eda217022718702bbebbbf50a92212361e54fbfa9795a508374e4f4f31939","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2024-11-01T14:57:16Z","title_canon_sha256":"986f5ba6fd4cc287cda3aa767e08afe4355d452bf75f797bbb8058e0576b8611"},"schema_version":"1.0","source":{"id":"2411.00640","kind":"arxiv","version":1}},"canonical_sha256":"824b5ca64bdb0054ac0fd03af146e5e645f3df353043df0bcd1903421b1d66fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"824b5ca64bdb0054ac0fd03af146e5e645f3df353043df0bcd1903421b1d66fd","first_computed_at":"2026-07-05T09:29:46.058980Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:29:46.058980Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dAnAScsWYs0wJ/OhNl1XRAOCghbCN387SvqprAO2zzN28P7RCbs5zvlTHaG63iLAwpKTr/FgYcqNTfGR8742Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:29:46.059457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.00640","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dce7700d5acd56878ecf1803e575389ed8db7ec4e6af81a09c21fbfc45af056e","sha256:76750dfc15db7905afa606b3f0e912fc7794f7e725a364dfaaa1ee9bde06bea7"],"state_sha256":"a5a285fd4231c81e1806a3fe6935af4d51d0c0f661c8d36caa137bb8dc170084"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cWnHmn4U7l4VScghmFn2zww+7K9678H3nOkw7hVSwR1/ydjR8NXPbBX5UfeEMwVG3+zzlbMg0UXy6SiDRiaaAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-10T01:41:58.225668Z","bundle_sha256":"d71613a5714dee55d50b099d63abed2d7e855baf5150d8ae9b0475fd3d748258"}}