{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:5JGTY6ORTJ4X6LJ6ZV3LBUPPI5","short_pith_number":"pith:5JGTY6OR","canonical_record":{"source":{"id":"2203.04444","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2022-03-08T23:31:06Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"271fcd8cfe7d665e8c4cd225e7e6cc535458317810765e253e4880e6208f58b9","abstract_canon_sha256":"c5210459b77ebf01180c8b886b88f4141b8ba3c8f26619ec7a373ccaec0be3e0"},"schema_version":"1.0"},"canonical_sha256":"ea4d3c79d19a797f2d3ecd76b0d1ef47603e4e07a28ca201b9e749cf6560e05d","source":{"kind":"arxiv","id":"2203.04444","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.04444","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"arxiv_version","alias_value":"2203.04444v1","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.04444","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_12","alias_value":"5JGTY6ORTJ4X","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_16","alias_value":"5JGTY6ORTJ4X6LJ6","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_8","alias_value":"5JGTY6OR","created_at":"2026-07-05T04:03:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:5JGTY6ORTJ4X6LJ6ZV3LBUPPI5","target":"record","payload":{"canonical_record":{"source":{"id":"2203.04444","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2022-03-08T23:31:06Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"271fcd8cfe7d665e8c4cd225e7e6cc535458317810765e253e4880e6208f58b9","abstract_canon_sha256":"c5210459b77ebf01180c8b886b88f4141b8ba3c8f26619ec7a373ccaec0be3e0"},"schema_version":"1.0"},"canonical_sha256":"ea4d3c79d19a797f2d3ecd76b0d1ef47603e4e07a28ca201b9e749cf6560e05d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:03:18.442069Z","signature_b64":"M0+kRGPSmTPrgZ9Txb0JdRrxI3AxeT2B1DEZjkuvjdV3sk3rPR82x4Y0um1TrKJs++YX4y8fehr1MKz5zbvZCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea4d3c79d19a797f2d3ecd76b0d1ef47603e4e07a28ca201b9e749cf6560e05d","last_reissued_at":"2026-07-05T04:03:18.441730Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:03:18.441730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2203.04444","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":"2026-07-05T04:03:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yZx5FZBnfxw9vU8MAvphZ5juPn/3ova8/wDffmOAeBdq345nl34pS058uitbYtwHCDJb22gz6yTUkWB5G2/vBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:41:27.048229Z"},"content_sha256":"5c524a8d9f5645deb0f4a13f1717665704721107783e9c63b725ab306d040f32","schema_version":"1.0","event_id":"sha256:5c524a8d9f5645deb0f4a13f1717665704721107783e9c63b725ab306d040f32"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:5JGTY6ORTJ4X6LJ6ZV3LBUPPI5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reproducible Subjective Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.HC","authors_text":"Brian Tang, Bryan Pardo, Gefei Tan, Max Morrison","submitted_at":"2022-03-08T23:31:06Z","abstract_excerpt":"Human perceptual studies are the gold standard for the evaluation of many research tasks in machine learning, linguistics, and psychology. However, these studies require significant time and cost to perform. As a result, many researchers use objective measures that can correlate poorly with human evaluation. When subjective evaluations are performed, they are often not reported with sufficient detail to ensure reproducibility. We propose Reproducible Subjective Evaluation (ReSEval), an open-source framework for quickly deploying crowdsourced subjective evaluations directly from Python. ReSEval"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.04444","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/2203.04444/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":"2026-07-05T04:03:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bRI9Z97FsyZutPNhFPvb4moX1Ru4nL3luF/qaBEtYdSoeyXTKnZ8Snx9vDe4dZlGobLO/RDZBY5qushJh9+XAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:41:27.048905Z"},"content_sha256":"93a386b0eaaf3c88fdb242de75365727d498a98912d52eae6977a0cd0bee7884","schema_version":"1.0","event_id":"sha256:93a386b0eaaf3c88fdb242de75365727d498a98912d52eae6977a0cd0bee7884"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/bundle.json","state_url":"https://pith.science/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/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":"2026-08-05T15:41:27Z","links":{"resolver":"https://pith.science/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5","bundle":"https://pith.science/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/bundle.json","state":"https://pith.science/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5JGTY6ORTJ4X6LJ6ZV3LBUPPI5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:5JGTY6ORTJ4X6LJ6ZV3LBUPPI5","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":"c5210459b77ebf01180c8b886b88f4141b8ba3c8f26619ec7a373ccaec0be3e0","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2022-03-08T23:31:06Z","title_canon_sha256":"271fcd8cfe7d665e8c4cd225e7e6cc535458317810765e253e4880e6208f58b9"},"schema_version":"1.0","source":{"id":"2203.04444","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.04444","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"arxiv_version","alias_value":"2203.04444v1","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.04444","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_12","alias_value":"5JGTY6ORTJ4X","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_16","alias_value":"5JGTY6ORTJ4X6LJ6","created_at":"2026-07-05T04:03:18Z"},{"alias_kind":"pith_short_8","alias_value":"5JGTY6OR","created_at":"2026-07-05T04:03:18Z"}],"graph_snapshots":[{"event_id":"sha256:93a386b0eaaf3c88fdb242de75365727d498a98912d52eae6977a0cd0bee7884","target":"graph","created_at":"2026-07-05T04:03:18Z","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/2203.04444/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human perceptual studies are the gold standard for the evaluation of many research tasks in machine learning, linguistics, and psychology. However, these studies require significant time and cost to perform. As a result, many researchers use objective measures that can correlate poorly with human evaluation. When subjective evaluations are performed, they are often not reported with sufficient detail to ensure reproducibility. We propose Reproducible Subjective Evaluation (ReSEval), an open-source framework for quickly deploying crowdsourced subjective evaluations directly from Python. ReSEval","authors_text":"Brian Tang, Bryan Pardo, Gefei Tan, Max Morrison","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2022-03-08T23:31:06Z","title":"Reproducible Subjective Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.04444","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":"sha256:5c524a8d9f5645deb0f4a13f1717665704721107783e9c63b725ab306d040f32","target":"record","created_at":"2026-07-05T04:03:18Z","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":"c5210459b77ebf01180c8b886b88f4141b8ba3c8f26619ec7a373ccaec0be3e0","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2022-03-08T23:31:06Z","title_canon_sha256":"271fcd8cfe7d665e8c4cd225e7e6cc535458317810765e253e4880e6208f58b9"},"schema_version":"1.0","source":{"id":"2203.04444","kind":"arxiv","version":1}},"canonical_sha256":"ea4d3c79d19a797f2d3ecd76b0d1ef47603e4e07a28ca201b9e749cf6560e05d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ea4d3c79d19a797f2d3ecd76b0d1ef47603e4e07a28ca201b9e749cf6560e05d","first_computed_at":"2026-07-05T04:03:18.441730Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:03:18.441730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"M0+kRGPSmTPrgZ9Txb0JdRrxI3AxeT2B1DEZjkuvjdV3sk3rPR82x4Y0um1TrKJs++YX4y8fehr1MKz5zbvZCg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:03:18.442069Z","signed_message":"canonical_sha256_bytes"},"source_id":"2203.04444","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5c524a8d9f5645deb0f4a13f1717665704721107783e9c63b725ab306d040f32","sha256:93a386b0eaaf3c88fdb242de75365727d498a98912d52eae6977a0cd0bee7884"],"state_sha256":"6d5e6795d14326df049b22cbbb58b59fd2ad08200c2e04ceecf24652c3e85e69"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iTn3iGXnOfEgMEjZGTtNzxxekCd6q1PeNaIhxmKPDmznlfhg8pyeJ0YsrOTwF3nz2RpFbkLb415JGVfbR78dBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T15:41:27.052594Z","bundle_sha256":"2d9d072daea79f5c98164b7b764fba226621f8022914b846aad815c985329f30"}}