{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:FFYHLEST5KXVIU6IAU3LEACBQ6","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":"aeaf557f59b45b0fd67dffc61c0099840bf34a3f3042704763e8daa0eaeec7b1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-29T16:28:43Z","title_canon_sha256":"b88724cb1183a229a7ea0041f9aca79511f37c9fcfd3e89cfce0bb3991f70352"},"schema_version":"1.0","source":{"id":"2303.16835","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.16835","created_at":"2026-07-05T05:56:10Z"},{"alias_kind":"arxiv_version","alias_value":"2303.16835v1","created_at":"2026-07-05T05:56:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.16835","created_at":"2026-07-05T05:56:10Z"},{"alias_kind":"pith_short_12","alias_value":"FFYHLEST5KXV","created_at":"2026-07-05T05:56:10Z"},{"alias_kind":"pith_short_16","alias_value":"FFYHLEST5KXVIU6I","created_at":"2026-07-05T05:56:10Z"},{"alias_kind":"pith_short_8","alias_value":"FFYHLEST","created_at":"2026-07-05T05:56:10Z"}],"graph_snapshots":[{"event_id":"sha256:b8e40b0cda90506b29f2b12c47de6a939073b52e4e93fe1ab13d00950c903bc5","target":"graph","created_at":"2026-07-05T05:56:10Z","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/2303.16835/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a large-scale empirical investigation of the zero-shot learning phenomena in a specific recognizing textual entailment (RTE) task category, i.e. the automated mining of leaderboards for Empirical AI Research. The prior reported state-of-the-art models for leaderboards extraction formulated as an RTE task, in a non-zero-shot setting, are promising with above 90% reported performances. However, a central research question remains unexamined: did the models actually learn entailment? Thus, for the experiments in this paper, two prior reported state-of-the-art models are tested out-of-t","authors_text":"Jennifer D'Souza, Salomon Kabongo, S\\\"oren Auer","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-29T16:28:43Z","title":"Zero-shot Entailment of Leaderboards for Empirical AI Research"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.16835","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:cac3b9ab2109a4734f356717a67a7a9721123c14cb83044bca28930462117926","target":"record","created_at":"2026-07-05T05:56:10Z","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":"aeaf557f59b45b0fd67dffc61c0099840bf34a3f3042704763e8daa0eaeec7b1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-29T16:28:43Z","title_canon_sha256":"b88724cb1183a229a7ea0041f9aca79511f37c9fcfd3e89cfce0bb3991f70352"},"schema_version":"1.0","source":{"id":"2303.16835","kind":"arxiv","version":1}},"canonical_sha256":"2970759253eaaf5453c80536b200418786d978a32d2a04c85df2aeee96af9deb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2970759253eaaf5453c80536b200418786d978a32d2a04c85df2aeee96af9deb","first_computed_at":"2026-07-05T05:56:10.827010Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:56:10.827010Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nt5HArpgLSjbVhEmR1ndrHUH+14NerM7S3HAN9M7nDOjtuvLVvX8EVeSQl4EjIgFxYgUGMlT9IUjhxATeA2zCg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:56:10.827502Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.16835","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cac3b9ab2109a4734f356717a67a7a9721123c14cb83044bca28930462117926","sha256:b8e40b0cda90506b29f2b12c47de6a939073b52e4e93fe1ab13d00950c903bc5"],"state_sha256":"72a0c8571df90f026a519c10f2a7b1da11d231c818facedfeba1bf2663250717"}