{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:ZBVHAP7S3UNPLAOPM4PMC7HPL2","short_pith_number":"pith:ZBVHAP7S","canonical_record":{"source":{"id":"2202.07654","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-02-15T18:53:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f4c00de2777bf5d10ae88c21fe5ddce86487236795a59477891b495ab4807a20","abstract_canon_sha256":"19a9894713ddbb4341333791e6b7b0892d96e1282fab7375a02bd4ff66bb23e2"},"schema_version":"1.0"},"canonical_sha256":"c86a703ff2dd1af581cf671ec17cef5ea5d509e84366a18cfb0b472eeae00696","source":{"kind":"arxiv","id":"2202.07654","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2202.07654","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"arxiv_version","alias_value":"2202.07654v2","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.07654","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_12","alias_value":"ZBVHAP7S3UNP","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_16","alias_value":"ZBVHAP7S3UNPLAOP","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_8","alias_value":"ZBVHAP7S","created_at":"2026-07-05T05:10:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:ZBVHAP7S3UNPLAOPM4PMC7HPL2","target":"record","payload":{"canonical_record":{"source":{"id":"2202.07654","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-02-15T18:53:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f4c00de2777bf5d10ae88c21fe5ddce86487236795a59477891b495ab4807a20","abstract_canon_sha256":"19a9894713ddbb4341333791e6b7b0892d96e1282fab7375a02bd4ff66bb23e2"},"schema_version":"1.0"},"canonical_sha256":"c86a703ff2dd1af581cf671ec17cef5ea5d509e84366a18cfb0b472eeae00696","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:10:36.640995Z","signature_b64":"AvOBN+61gsmVAZxRkVV4jYk+Ic0akTnC1btRZbAlLyacP16GcFV7BIM+RtPUzfT1VZnhH6D1xs9le08o93h9Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c86a703ff2dd1af581cf671ec17cef5ea5d509e84366a18cfb0b472eeae00696","last_reissued_at":"2026-07-05T05:10:36.640514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:10:36.640514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2202.07654","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:10:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fmxhEoQMz25UfAnJ/fhit/PPWJ6ippxZakSHh5l9/qQwdzsQGblm9tEP5an7/L4T9fGmrAuqglvcJbBJpGuaCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T04:20:46.429739Z"},"content_sha256":"89d3cefe7a38030e6e1ead15e270d568e872b1f9387a8e7cdf2cba0e813caeca","schema_version":"1.0","event_id":"sha256:89d3cefe7a38030e6e1ead15e270d568e872b1f9387a8e7cdf2cba0e813caeca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:ZBVHAP7S3UNPLAOPM4PMC7HPL2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Benjamin Boerschinger, Christian Buck, Jannis Bulian, Tal Schuster, Wojciech Gajewski","submitted_at":"2022-02-15T18:53:58Z","abstract_excerpt":"The predictions of question answering (QA)systems are typically evaluated against manually annotated finite sets of one or more answers. This leads to a coverage limitation that results in underestimating the true performance of systems, and is typically addressed by extending over exact match (EM) with pre-defined rules or with the token-level F1 measure. In this paper, we present the first systematic conceptual and data-driven analysis to examine the shortcomings of token-level equivalence measures.\n  To this end, we define the asymmetric notion of answer equivalence (AE), accepting answers "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.07654","kind":"arxiv","version":2},"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/2202.07654/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":"2026-07-05T05:10:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"69w2954YE1+6kz1zWOFtOEbTwS9+4IJMCWehq6h8EWvY6KL2109sXIOFuKzr/Vu47FezA3MmRl9liN+pWSR/BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T04:20:46.430254Z"},"content_sha256":"48919b8e3c13ae7cd423a1796d593fc1e3988aedf0050f0f2f9928f7852a2d0e","schema_version":"1.0","event_id":"sha256:48919b8e3c13ae7cd423a1796d593fc1e3988aedf0050f0f2f9928f7852a2d0e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/bundle.json","state_url":"https://pith.science/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/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":"2026-08-07T04:20:46Z","links":{"resolver":"https://pith.science/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2","bundle":"https://pith.science/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/bundle.json","state":"https://pith.science/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZBVHAP7S3UNPLAOPM4PMC7HPL2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:ZBVHAP7S3UNPLAOPM4PMC7HPL2","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":"19a9894713ddbb4341333791e6b7b0892d96e1282fab7375a02bd4ff66bb23e2","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-02-15T18:53:58Z","title_canon_sha256":"f4c00de2777bf5d10ae88c21fe5ddce86487236795a59477891b495ab4807a20"},"schema_version":"1.0","source":{"id":"2202.07654","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2202.07654","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"arxiv_version","alias_value":"2202.07654v2","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.07654","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_12","alias_value":"ZBVHAP7S3UNP","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_16","alias_value":"ZBVHAP7S3UNPLAOP","created_at":"2026-07-05T05:10:36Z"},{"alias_kind":"pith_short_8","alias_value":"ZBVHAP7S","created_at":"2026-07-05T05:10:36Z"}],"graph_snapshots":[{"event_id":"sha256:48919b8e3c13ae7cd423a1796d593fc1e3988aedf0050f0f2f9928f7852a2d0e","target":"graph","created_at":"2026-07-05T05:10:36Z","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/2202.07654/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The predictions of question answering (QA)systems are typically evaluated against manually annotated finite sets of one or more answers. This leads to a coverage limitation that results in underestimating the true performance of systems, and is typically addressed by extending over exact match (EM) with pre-defined rules or with the token-level F1 measure. In this paper, we present the first systematic conceptual and data-driven analysis to examine the shortcomings of token-level equivalence measures.\n  To this end, we define the asymmetric notion of answer equivalence (AE), accepting answers ","authors_text":"Benjamin Boerschinger, Christian Buck, Jannis Bulian, Tal Schuster, Wojciech Gajewski","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-02-15T18:53:58Z","title":"Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.07654","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":"sha256:89d3cefe7a38030e6e1ead15e270d568e872b1f9387a8e7cdf2cba0e813caeca","target":"record","created_at":"2026-07-05T05:10:36Z","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":"19a9894713ddbb4341333791e6b7b0892d96e1282fab7375a02bd4ff66bb23e2","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-02-15T18:53:58Z","title_canon_sha256":"f4c00de2777bf5d10ae88c21fe5ddce86487236795a59477891b495ab4807a20"},"schema_version":"1.0","source":{"id":"2202.07654","kind":"arxiv","version":2}},"canonical_sha256":"c86a703ff2dd1af581cf671ec17cef5ea5d509e84366a18cfb0b472eeae00696","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c86a703ff2dd1af581cf671ec17cef5ea5d509e84366a18cfb0b472eeae00696","first_computed_at":"2026-07-05T05:10:36.640514Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:10:36.640514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AvOBN+61gsmVAZxRkVV4jYk+Ic0akTnC1btRZbAlLyacP16GcFV7BIM+RtPUzfT1VZnhH6D1xs9le08o93h9Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:10:36.640995Z","signed_message":"canonical_sha256_bytes"},"source_id":"2202.07654","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:89d3cefe7a38030e6e1ead15e270d568e872b1f9387a8e7cdf2cba0e813caeca","sha256:48919b8e3c13ae7cd423a1796d593fc1e3988aedf0050f0f2f9928f7852a2d0e"],"state_sha256":"93f1c211dc19e6a2fbd9ba4b2c6e39a7dbdee99946c1c325c82742352f819a80"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2vULkKVrNYIao8gTHpWgCOUA7lw3qwSPMia0JKn30zSaN1UKlCxbicfgE2a7EOAl8xp6I5O9rH5+65GMyn8BCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T04:20:46.433939Z","bundle_sha256":"b3031ec55423c8cd2fbec813cfd312965c5c5d8a4ad471edab4019de6e6e2f0c"}}