{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MM66BDXCM6NL5YG2WI7TZ6BLCA","short_pith_number":"pith:MM66BDXC","schema_version":"1.0","canonical_sha256":"633de08ee2679abee0dab23f3cf82b10001f15236bb571c472b6384042c7f934","source":{"kind":"arxiv","id":"2203.12865","version":3},"attestation_state":"computed","paper":{"title":"Multilingual CheckList: Generation and Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Karthikeyan K, Monojit Choudhury, Pankaj Singh, Sandipan Dandapat, Shaily Bhatt, Somak Aditya, Sunayana Sitaram","submitted_at":"2022-03-24T06:05:28Z","abstract_excerpt":"Multilingual evaluation benchmarks usually contain limited high-resource languages and do not test models for specific linguistic capabilities. CheckList is a template-based evaluation approach that tests models for specific capabilities. The CheckList template creation process requires native speakers, posing a challenge in scaling to hundreds of languages. In this work, we explore multiple approaches to generate Multilingual CheckLists. We device an algorithm - Template Extraction Algorithm (TEA) for automatically extracting target language CheckList templates from machine translated instanc"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2203.12865","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-03-24T06:05:28Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"348288871991ac9992e6c5cd9985ac3445e0293df6df827f2589211879cb8555","abstract_canon_sha256":"5cb847c1032634d7633bf56c28a53b3775f139ac3ec953733017ffeaab0893ee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:46.200742Z","signature_b64":"I9QrFD7UUhzAm2goF7mFP5N8VyBESP6d2/QY7v2Em2zC2ahBDcz7f6T/khYFH9GwoW1PBiwJwX5RT2ypjXQQBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"633de08ee2679abee0dab23f3cf82b10001f15236bb571c472b6384042c7f934","last_reissued_at":"2026-07-05T05:05:46.200286Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:46.200286Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual CheckList: Generation and Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Karthikeyan K, Monojit Choudhury, Pankaj Singh, Sandipan Dandapat, Shaily Bhatt, Somak Aditya, Sunayana Sitaram","submitted_at":"2022-03-24T06:05:28Z","abstract_excerpt":"Multilingual evaluation benchmarks usually contain limited high-resource languages and do not test models for specific linguistic capabilities. CheckList is a template-based evaluation approach that tests models for specific capabilities. The CheckList template creation process requires native speakers, posing a challenge in scaling to hundreds of languages. In this work, we explore multiple approaches to generate Multilingual CheckLists. We device an algorithm - Template Extraction Algorithm (TEA) for automatically extracting target language CheckList templates from machine translated instanc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.12865","kind":"arxiv","version":3},"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/2203.12865/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2203.12865","created_at":"2026-07-05T05:05:46.200352+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.12865v3","created_at":"2026-07-05T05:05:46.200352+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.12865","created_at":"2026-07-05T05:05:46.200352+00:00"},{"alias_kind":"pith_short_12","alias_value":"MM66BDXCM6NL","created_at":"2026-07-05T05:05:46.200352+00:00"},{"alias_kind":"pith_short_16","alias_value":"MM66BDXCM6NL5YG2","created_at":"2026-07-05T05:05:46.200352+00:00"},{"alias_kind":"pith_short_8","alias_value":"MM66BDXC","created_at":"2026-07-05T05:05:46.200352+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA","json":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA.json","graph_json":"https://pith.science/api/pith-number/MM66BDXCM6NL5YG2WI7TZ6BLCA/graph.json","events_json":"https://pith.science/api/pith-number/MM66BDXCM6NL5YG2WI7TZ6BLCA/events.json","paper":"https://pith.science/paper/MM66BDXC"},"agent_actions":{"view_html":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA","download_json":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA.json","view_paper":"https://pith.science/paper/MM66BDXC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.12865&json=true","fetch_graph":"https://pith.science/api/pith-number/MM66BDXCM6NL5YG2WI7TZ6BLCA/graph.json","fetch_events":"https://pith.science/api/pith-number/MM66BDXCM6NL5YG2WI7TZ6BLCA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA/action/storage_attestation","attest_author":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA/action/author_attestation","sign_citation":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA/action/citation_signature","submit_replication":"https://pith.science/pith/MM66BDXCM6NL5YG2WI7TZ6BLCA/action/replication_record"}},"created_at":"2026-07-05T05:05:46.200352+00:00","updated_at":"2026-07-05T05:05:46.200352+00:00"}