{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Q767BO3NVILZND3LHG65YCN44N","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":"80090b7ceca213cbfaaa22ae5dc73f6bfef4cfceef539a642002c14076c3c3f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T06:27:47Z","title_canon_sha256":"da31e93d9b673680ff59a3d7f677e5a4e9b71fdf87b1ec3d806d01545e17ef5f"},"schema_version":"1.0","source":{"id":"2406.16342","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.16342","created_at":"2026-07-05T10:16:34Z"},{"alias_kind":"arxiv_version","alias_value":"2406.16342v3","created_at":"2026-07-05T10:16:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.16342","created_at":"2026-07-05T10:16:34Z"},{"alias_kind":"pith_short_12","alias_value":"Q767BO3NVILZ","created_at":"2026-07-05T10:16:34Z"},{"alias_kind":"pith_short_16","alias_value":"Q767BO3NVILZND3L","created_at":"2026-07-05T10:16:34Z"},{"alias_kind":"pith_short_8","alias_value":"Q767BO3N","created_at":"2026-07-05T10:16:34Z"}],"graph_snapshots":[{"event_id":"sha256:d7f4b85669eb91777d3e502dcd0f269958615bc3b3c891d1a282f42fafab46a3","target":"graph","created_at":"2026-07-05T10:16:34Z","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/2406.16342/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adversarial datasets should validate AI robustness by providing samples on which humans perform well, but models do not. However, as models evolve, datasets can become obsolete. Measuring whether a dataset remains adversarial is hindered by the lack of a standardized metric for measuring adversarialness. We propose AdvScore, a human-grounded evaluation metric that assesses a dataset's adversarialness by capturing models' and humans' varying abilities while also identifying poor examples. We then use AdvScore to motivate a new dataset creation pipeline for realistic and high-quality adversarial","authors_text":"Eve Fleisig, Ishani Mondal, Jordan Lee Boyd-Graber, Maharshi Gor, Yoo Yeon Sung","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T06:27:47Z","title":"Is your benchmark truly adversarial? AdvScore: Evaluating Human-Grounded Adversarialness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.16342","kind":"arxiv","version":3},"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:b931c2758a1e00617b3f68f9e822fd4540a9cdb530b7b673a14867b3e14ede77","target":"record","created_at":"2026-07-05T10:16:34Z","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":"80090b7ceca213cbfaaa22ae5dc73f6bfef4cfceef539a642002c14076c3c3f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T06:27:47Z","title_canon_sha256":"da31e93d9b673680ff59a3d7f677e5a4e9b71fdf87b1ec3d806d01545e17ef5f"},"schema_version":"1.0","source":{"id":"2406.16342","kind":"arxiv","version":3}},"canonical_sha256":"87fdf0bb6daa17968f6b39bddc09bce3710b0107ada01d325cb4cd3243acd38b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87fdf0bb6daa17968f6b39bddc09bce3710b0107ada01d325cb4cd3243acd38b","first_computed_at":"2026-07-05T10:16:34.522119Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:16:34.522119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nDI/4m4bO+TEweIQZYXesN8DqND7eN2nLF+p3LKeumpy452IQg5K/CGXLo1IWxJNgIl3KBN2rovNff/wtoauAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:16:34.522601Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.16342","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b931c2758a1e00617b3f68f9e822fd4540a9cdb530b7b673a14867b3e14ede77","sha256:d7f4b85669eb91777d3e502dcd0f269958615bc3b3c891d1a282f42fafab46a3"],"state_sha256":"d2da622417414e0c4bf5c66ca64b693b433162a5b74098af6f1b02027220c855"}