{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:55Y7NJEJ424ZW5DX3OTOLU4PNV","short_pith_number":"pith:55Y7NJEJ","schema_version":"1.0","canonical_sha256":"ef71f6a489e6b99b7477dba6e5d38f6d7521900befd48b888cfb36a6a57d37a9","source":{"kind":"arxiv","id":"2210.01734","version":1},"attestation_state":"computed","paper":{"title":"Text Characterization Toolkit","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Daniel Simig, Dieuwke Hupkes, Khuyagbaatar Batsuren, Mona Diab, Peter Henderson, Tianlu Wang, Verna Dankers","submitted_at":"2022-10-04T16:54:11Z","abstract_excerpt":"In NLP, models are usually evaluated by reporting single-number performance scores on a number of readily available benchmarks, without much deeper analysis. Here, we argue that - especially given the well-known fact that benchmarks often contain biases, artefacts, and spurious correlations - deeper results analysis should become the de-facto standard when presenting new models or benchmarks. We present a tool that researchers can use to study properties of the dataset and the influence of those properties on their models' behaviour. Our Text Characterization Toolkit includes both an easy-to-u"},"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":"2210.01734","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-04T16:54:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a4f854ddc4f830db1b8cd58c947cb521a8f267f61d682b4cad30980fc132d5b6","abstract_canon_sha256":"5864fcfe72f82c3799a5ffa8154a72f6f64b9ad56493dcef2887b1e9750411e8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:03:13.566136Z","signature_b64":"T110c4fgVo//fhGPwV8/Ri/pKKTsfsYKHtlwo4xI2eesW6ectvfqYK23HFJMyO4xaUYDF4MZaJFCghXTvynZCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef71f6a489e6b99b7477dba6e5d38f6d7521900befd48b888cfb36a6a57d37a9","last_reissued_at":"2026-07-05T05:03:13.565747Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:03:13.565747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Text Characterization Toolkit","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Daniel Simig, Dieuwke Hupkes, Khuyagbaatar Batsuren, Mona Diab, Peter Henderson, Tianlu Wang, Verna Dankers","submitted_at":"2022-10-04T16:54:11Z","abstract_excerpt":"In NLP, models are usually evaluated by reporting single-number performance scores on a number of readily available benchmarks, without much deeper analysis. Here, we argue that - especially given the well-known fact that benchmarks often contain biases, artefacts, and spurious correlations - deeper results analysis should become the de-facto standard when presenting new models or benchmarks. We present a tool that researchers can use to study properties of the dataset and the influence of those properties on their models' behaviour. Our Text Characterization Toolkit includes both an easy-to-u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.01734","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/2210.01734/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":"2210.01734","created_at":"2026-07-05T05:03:13.565803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.01734v1","created_at":"2026-07-05T05:03:13.565803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.01734","created_at":"2026-07-05T05:03:13.565803+00:00"},{"alias_kind":"pith_short_12","alias_value":"55Y7NJEJ424Z","created_at":"2026-07-05T05:03:13.565803+00:00"},{"alias_kind":"pith_short_16","alias_value":"55Y7NJEJ424ZW5DX","created_at":"2026-07-05T05:03:13.565803+00:00"},{"alias_kind":"pith_short_8","alias_value":"55Y7NJEJ","created_at":"2026-07-05T05:03:13.565803+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/55Y7NJEJ424ZW5DX3OTOLU4PNV","json":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV.json","graph_json":"https://pith.science/api/pith-number/55Y7NJEJ424ZW5DX3OTOLU4PNV/graph.json","events_json":"https://pith.science/api/pith-number/55Y7NJEJ424ZW5DX3OTOLU4PNV/events.json","paper":"https://pith.science/paper/55Y7NJEJ"},"agent_actions":{"view_html":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV","download_json":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV.json","view_paper":"https://pith.science/paper/55Y7NJEJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.01734&json=true","fetch_graph":"https://pith.science/api/pith-number/55Y7NJEJ424ZW5DX3OTOLU4PNV/graph.json","fetch_events":"https://pith.science/api/pith-number/55Y7NJEJ424ZW5DX3OTOLU4PNV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV/action/storage_attestation","attest_author":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV/action/author_attestation","sign_citation":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV/action/citation_signature","submit_replication":"https://pith.science/pith/55Y7NJEJ424ZW5DX3OTOLU4PNV/action/replication_record"}},"created_at":"2026-07-05T05:03:13.565803+00:00","updated_at":"2026-07-05T05:03:13.565803+00:00"}