{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:EYNHOD27SOCK4JMYSD63SOKRCT","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":"737d8d9816e82c890db6eb32432ab8b77a6772849ed63134c18a6ae40a9ae8a1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-06T15:29:57Z","title_canon_sha256":"de91f2a0f96d72b0b102389ce017ec3a9a4209c64c671fbe8e0d46ee3338c7b4"},"schema_version":"1.0","source":{"id":"2304.03145","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.03145","created_at":"2026-07-05T08:08:48Z"},{"alias_kind":"arxiv_version","alias_value":"2304.03145v2","created_at":"2026-07-05T08:08:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.03145","created_at":"2026-07-05T08:08:48Z"},{"alias_kind":"pith_short_12","alias_value":"EYNHOD27SOCK","created_at":"2026-07-05T08:08:48Z"},{"alias_kind":"pith_short_16","alias_value":"EYNHOD27SOCK4JMY","created_at":"2026-07-05T08:08:48Z"},{"alias_kind":"pith_short_8","alias_value":"EYNHOD27","created_at":"2026-07-05T08:08:48Z"}],"graph_snapshots":[{"event_id":"sha256:d770c4e46836ed5b888493af1c4acaddf67e8ad2174b84c01be1cf560d37eed5","target":"graph","created_at":"2026-07-05T08:08:48Z","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/2304.03145/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Question answering (QA) models have shown compelling results in the task of Machine Reading Comprehension (MRC). Recently these systems have proved to perform better than humans on held-out test sets of datasets e.g. SQuAD, but their robustness is not guaranteed. The QA model's brittleness is exposed when evaluated on adversarial generated examples by a performance drop. In this study, we explore the robustness of MRC models to entity renaming, with entities from low-resource regions such as Africa. We propose EntSwap, a method for test-time perturbations, to create a test set whose entities h","authors_text":"Clemencia Siro, Tunde Oluwaseyi Ajayi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-06T15:29:57Z","title":"Evaluating the Robustness of Machine Reading Comprehension Models to Low Resource Entity Renaming"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.03145","kind":"arxiv","version":2},"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:c4ede571620f9a1dfefe28054d1c663cbe63b185c1e38d3ac69457558713b762","target":"record","created_at":"2026-07-05T08:08:48Z","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":"737d8d9816e82c890db6eb32432ab8b77a6772849ed63134c18a6ae40a9ae8a1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-06T15:29:57Z","title_canon_sha256":"de91f2a0f96d72b0b102389ce017ec3a9a4209c64c671fbe8e0d46ee3338c7b4"},"schema_version":"1.0","source":{"id":"2304.03145","kind":"arxiv","version":2}},"canonical_sha256":"261a770f5f9384ae259890fdb9395114f8df93f162b2f64073bf6cca771aeedf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"261a770f5f9384ae259890fdb9395114f8df93f162b2f64073bf6cca771aeedf","first_computed_at":"2026-07-05T08:08:48.206399Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:08:48.206399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iVNAF30/bD1h+2ly1sG/dr1VtjkH8pPvjXPPlLxQMpcNpwUNwTLEvZdFAuZOJjrcdVukrut9G63JZUGozJogCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:08:48.206848Z","signed_message":"canonical_sha256_bytes"},"source_id":"2304.03145","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c4ede571620f9a1dfefe28054d1c663cbe63b185c1e38d3ac69457558713b762","sha256:d770c4e46836ed5b888493af1c4acaddf67e8ad2174b84c01be1cf560d37eed5"],"state_sha256":"13a5aeff21301386f081aa04c6982f7a17436a720d47827324b5e3d89cf0f093"}