{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TIBGYRW2O4Z5752YKOAGOMJE4O","short_pith_number":"pith:TIBGYRW2","canonical_record":{"source":{"id":"2506.07949","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T17:14:41Z","cross_cats_sorted":[],"title_canon_sha256":"ee713f322985cb177b073e149cfdaa7b6b229f2f596bb7ef43b6a7fc9c94cb72","abstract_canon_sha256":"1a87397f70e4c73f9b384b54f081ac1efd7a2fc896094351434a395b6b7cc11c"},"schema_version":"1.0"},"canonical_sha256":"9a026c46da7733dff7585380673124e393edd6bbba36e06ade8bb87014fef2ba","source":{"kind":"arxiv","id":"2506.07949","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.07949","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"arxiv_version","alias_value":"2506.07949v1","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07949","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_12","alias_value":"TIBGYRW2O4Z5","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_16","alias_value":"TIBGYRW2O4Z5752Y","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_8","alias_value":"TIBGYRW2","created_at":"2026-07-05T11:18:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TIBGYRW2O4Z5752YKOAGOMJE4O","target":"record","payload":{"canonical_record":{"source":{"id":"2506.07949","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T17:14:41Z","cross_cats_sorted":[],"title_canon_sha256":"ee713f322985cb177b073e149cfdaa7b6b229f2f596bb7ef43b6a7fc9c94cb72","abstract_canon_sha256":"1a87397f70e4c73f9b384b54f081ac1efd7a2fc896094351434a395b6b7cc11c"},"schema_version":"1.0"},"canonical_sha256":"9a026c46da7733dff7585380673124e393edd6bbba36e06ade8bb87014fef2ba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:35.936974Z","signature_b64":"0R4OXQuJA8TY6hi6hmWyx9RqwPCKJXzPKUmcR1BCgfNHLsz84EMskleuizVD/eCoCm5yFSkGHXPJ5PPVezDnCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a026c46da7733dff7585380673124e393edd6bbba36e06ade8bb87014fef2ba","last_reissued_at":"2026-07-05T11:18:35.936447Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:35.936447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.07949","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":"2026-07-05T11:18:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y5SiDyRZOGuFiY7vSTWzli1n5ydDnDjs+LZF2RflTk8oa2OEYggBUiaSTIWnf+VFtTBgTbyyL7cbhv7psCMACw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T23:36:32.600571Z"},"content_sha256":"00ddf1a6022f1d6c664928407da490ff9e06ab18f31daa26999b2d48cd0dbe25","schema_version":"1.0","event_id":"sha256:00ddf1a6022f1d6c664928407da490ff9e06ab18f31daa26999b2d48cd0dbe25"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TIBGYRW2O4Z5752YKOAGOMJE4O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cost-Optimal Active AI Model Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Adam Fisch, Alekh Agarwal, Anastasios N. Angelopoulos, Jacob Eisenstein, Jonathan Berant","submitted_at":"2025-06-09T17:14:41Z","abstract_excerpt":"The development lifecycle of generative AI systems requires continual evaluation, data acquisition, and annotation, which is costly in both resources and time. In practice, rapid iteration often makes it necessary to rely on synthetic annotation data because of the low cost, despite the potential for substantial bias. In this paper, we develop novel, cost-aware methods for actively balancing the use of a cheap, but often inaccurate, weak rater -- such as a model-based autorater that is designed to automatically assess the quality of generated content -- with a more expensive, but also more acc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07949","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/2506.07949/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":"2026-07-05T11:18:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IrtWhEoLcV7LBStMjbXqs2YEgXjb78uoeiyBzSnPJNLrBfEyNbGUHc3lHPYebdtPYpcT9hPWqCbaYNwH5FyJCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T23:36:32.601468Z"},"content_sha256":"fd5bff3263f41e88040e970280036415ef6be62a163c593a31279e08f22da722","schema_version":"1.0","event_id":"sha256:fd5bff3263f41e88040e970280036415ef6be62a163c593a31279e08f22da722"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/bundle.json","state_url":"https://pith.science/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/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":"2026-08-04T23:36:32Z","links":{"resolver":"https://pith.science/pith/TIBGYRW2O4Z5752YKOAGOMJE4O","bundle":"https://pith.science/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/bundle.json","state":"https://pith.science/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TIBGYRW2O4Z5752YKOAGOMJE4O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TIBGYRW2O4Z5752YKOAGOMJE4O","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":"1a87397f70e4c73f9b384b54f081ac1efd7a2fc896094351434a395b6b7cc11c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T17:14:41Z","title_canon_sha256":"ee713f322985cb177b073e149cfdaa7b6b229f2f596bb7ef43b6a7fc9c94cb72"},"schema_version":"1.0","source":{"id":"2506.07949","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.07949","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"arxiv_version","alias_value":"2506.07949v1","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07949","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_12","alias_value":"TIBGYRW2O4Z5","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_16","alias_value":"TIBGYRW2O4Z5752Y","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_8","alias_value":"TIBGYRW2","created_at":"2026-07-05T11:18:35Z"}],"graph_snapshots":[{"event_id":"sha256:fd5bff3263f41e88040e970280036415ef6be62a163c593a31279e08f22da722","target":"graph","created_at":"2026-07-05T11:18:35Z","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/2506.07949/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The development lifecycle of generative AI systems requires continual evaluation, data acquisition, and annotation, which is costly in both resources and time. In practice, rapid iteration often makes it necessary to rely on synthetic annotation data because of the low cost, despite the potential for substantial bias. In this paper, we develop novel, cost-aware methods for actively balancing the use of a cheap, but often inaccurate, weak rater -- such as a model-based autorater that is designed to automatically assess the quality of generated content -- with a more expensive, but also more acc","authors_text":"Adam Fisch, Alekh Agarwal, Anastasios N. Angelopoulos, Jacob Eisenstein, Jonathan Berant","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T17:14:41Z","title":"Cost-Optimal Active AI Model Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07949","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":"sha256:00ddf1a6022f1d6c664928407da490ff9e06ab18f31daa26999b2d48cd0dbe25","target":"record","created_at":"2026-07-05T11:18:35Z","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":"1a87397f70e4c73f9b384b54f081ac1efd7a2fc896094351434a395b6b7cc11c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T17:14:41Z","title_canon_sha256":"ee713f322985cb177b073e149cfdaa7b6b229f2f596bb7ef43b6a7fc9c94cb72"},"schema_version":"1.0","source":{"id":"2506.07949","kind":"arxiv","version":1}},"canonical_sha256":"9a026c46da7733dff7585380673124e393edd6bbba36e06ade8bb87014fef2ba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9a026c46da7733dff7585380673124e393edd6bbba36e06ade8bb87014fef2ba","first_computed_at":"2026-07-05T11:18:35.936447Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:35.936447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0R4OXQuJA8TY6hi6hmWyx9RqwPCKJXzPKUmcR1BCgfNHLsz84EMskleuizVD/eCoCm5yFSkGHXPJ5PPVezDnCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:35.936974Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.07949","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:00ddf1a6022f1d6c664928407da490ff9e06ab18f31daa26999b2d48cd0dbe25","sha256:fd5bff3263f41e88040e970280036415ef6be62a163c593a31279e08f22da722"],"state_sha256":"066b0a896edc95f46dcb505ecdb600f4b7ed1d47ef16f525792d4cc5e443df37"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T/7PoS4hzTtL2oefFARXGyEFfVPUZ/UAIR2ZZ1xrlOjVu2luGyfmfeVzrzPXaiseI0qoBfU2414+uYQW7QNZDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T23:36:32.606875Z","bundle_sha256":"dae7bf5e08f76457ce1b68fdedaa7ccb7f5b82488e13d2b849744d58230a28a5"}}