{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:V4JYP6P3ZLOEMVVMYJD2MRAO2G","short_pith_number":"pith:V4JYP6P3","schema_version":"1.0","canonical_sha256":"af1387f9fbcadc4656acc247a6440ed1ab50ab150561a16c09b6133aaf7aa3b0","source":{"kind":"arxiv","id":"2211.07534","version":1},"attestation_state":"computed","paper":{"title":"High-Resource Methodological Bias in Low-Resource Investigations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Grangier, Maartje ter Hoeve, Natalie Schluter","submitted_at":"2022-11-14T17:04:38Z","abstract_excerpt":"The central bottleneck for low-resource NLP is typically regarded to be the quantity of accessible data, overlooking the contribution of data quality. This is particularly seen in the development and evaluation of low-resource systems via down sampling of high-resource language data. In this work we investigate the validity of this approach, and we specifically focus on two well-known NLP tasks for our empirical investigations: POS-tagging and machine translation. We show that down sampling from a high-resource language results in datasets with different properties than the low-resource datase"},"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":"2211.07534","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-11-14T17:04:38Z","cross_cats_sorted":[],"title_canon_sha256":"2212b3ae0ace66b3d2c8caaf19e975f39939ca57c8191dac69b1b2fbd71e990a","abstract_canon_sha256":"b3f1fcde3a0b31facd163f53ba19c9fcab181c47ab2ad08e8d0535659a20e6a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:15:54.732425Z","signature_b64":"C/bwPuPmjcRI+lZjh6zSh3ez2yGVhuyk/YV/DPPQcME2/zkTribjsZp7YZjOBr8fcYtO2kxGhEmAFJ77y3/tAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af1387f9fbcadc4656acc247a6440ed1ab50ab150561a16c09b6133aaf7aa3b0","last_reissued_at":"2026-07-05T05:15:54.731602Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:15:54.731602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"High-Resource Methodological Bias in Low-Resource Investigations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Grangier, Maartje ter Hoeve, Natalie Schluter","submitted_at":"2022-11-14T17:04:38Z","abstract_excerpt":"The central bottleneck for low-resource NLP is typically regarded to be the quantity of accessible data, overlooking the contribution of data quality. This is particularly seen in the development and evaluation of low-resource systems via down sampling of high-resource language data. In this work we investigate the validity of this approach, and we specifically focus on two well-known NLP tasks for our empirical investigations: POS-tagging and machine translation. We show that down sampling from a high-resource language results in datasets with different properties than the low-resource datase"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.07534","kind":"arxiv","version":1},"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/2211.07534/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":"2211.07534","created_at":"2026-07-05T05:15:54.731657+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.07534v1","created_at":"2026-07-05T05:15:54.731657+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.07534","created_at":"2026-07-05T05:15:54.731657+00:00"},{"alias_kind":"pith_short_12","alias_value":"V4JYP6P3ZLOE","created_at":"2026-07-05T05:15:54.731657+00:00"},{"alias_kind":"pith_short_16","alias_value":"V4JYP6P3ZLOEMVVM","created_at":"2026-07-05T05:15:54.731657+00:00"},{"alias_kind":"pith_short_8","alias_value":"V4JYP6P3","created_at":"2026-07-05T05:15:54.731657+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.04497","citing_title":"Opportunities and Challenges of Large Language Models for Low-Resource Languages in Humanities Research","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G","json":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G.json","graph_json":"https://pith.science/api/pith-number/V4JYP6P3ZLOEMVVMYJD2MRAO2G/graph.json","events_json":"https://pith.science/api/pith-number/V4JYP6P3ZLOEMVVMYJD2MRAO2G/events.json","paper":"https://pith.science/paper/V4JYP6P3"},"agent_actions":{"view_html":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G","download_json":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G.json","view_paper":"https://pith.science/paper/V4JYP6P3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.07534&json=true","fetch_graph":"https://pith.science/api/pith-number/V4JYP6P3ZLOEMVVMYJD2MRAO2G/graph.json","fetch_events":"https://pith.science/api/pith-number/V4JYP6P3ZLOEMVVMYJD2MRAO2G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G/action/storage_attestation","attest_author":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G/action/author_attestation","sign_citation":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G/action/citation_signature","submit_replication":"https://pith.science/pith/V4JYP6P3ZLOEMVVMYJD2MRAO2G/action/replication_record"}},"created_at":"2026-07-05T05:15:54.731657+00:00","updated_at":"2026-07-05T05:15:54.731657+00:00"}