{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:GKKKN2GZP23IE5UKWRYL3NCGX4","short_pith_number":"pith:GKKKN2GZ","schema_version":"1.0","canonical_sha256":"3294a6e8d97eb682768ab470bdb446bf11f61b3f40d0b91cf91f80582f536fa8","source":{"kind":"arxiv","id":"1803.09578","version":1},"attestation_state":"computed","paper":{"title":"Why Comparing Single Performance Scores Does Not Allow to Draw Conclusions About Machine Learning Approaches","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Iryna Gurevych, Nils Reimers","submitted_at":"2018-03-26T13:35:14Z","abstract_excerpt":"Developing state-of-the-art approaches for specific tasks is a major driving force in our research community. Depending on the prestige of the task, publishing it can come along with a lot of visibility. The question arises how reliable are our evaluation methodologies to compare approaches?\n  One common methodology to identify the state-of-the-art is to partition data into a train, a development and a test set. Researchers can train and tune their approach on some part of the dataset and then select the model that worked best on the development set for a final evaluation on unseen test data. "},"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":"1803.09578","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2018-03-26T13:35:14Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"title_canon_sha256":"8435b2f05420017354153bb3288e159ae301ca6905844d1f40c8ec86cd5c9890","abstract_canon_sha256":"bdcf8d2805e54c091688f472bdc07ccc32f3a278ff39ffa8bb7e66fc27d4e4f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:20:11.057130Z","signature_b64":"fpwKTcWRDnhs+QpiTYdWuqyx1KfHJHKa74Zg/5uPecB2A/akuS+MYfXzALHreBTuIvCZi/0PxmEmPBkB4LYSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3294a6e8d97eb682768ab470bdb446bf11f61b3f40d0b91cf91f80582f536fa8","last_reissued_at":"2026-05-18T00:20:11.056433Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:20:11.056433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why Comparing Single Performance Scores Does Not Allow to Draw Conclusions About Machine Learning Approaches","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Iryna Gurevych, Nils Reimers","submitted_at":"2018-03-26T13:35:14Z","abstract_excerpt":"Developing state-of-the-art approaches for specific tasks is a major driving force in our research community. Depending on the prestige of the task, publishing it can come along with a lot of visibility. The question arises how reliable are our evaluation methodologies to compare approaches?\n  One common methodology to identify the state-of-the-art is to partition data into a train, a development and a test set. Researchers can train and tune their approach on some part of the dataset and then select the model that worked best on the development set for a final evaluation on unseen test data. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.09578","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":""},"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":"1803.09578","created_at":"2026-05-18T00:20:11.056547+00:00"},{"alias_kind":"arxiv_version","alias_value":"1803.09578v1","created_at":"2026-05-18T00:20:11.056547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.09578","created_at":"2026-05-18T00:20:11.056547+00:00"},{"alias_kind":"pith_short_12","alias_value":"GKKKN2GZP23I","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_16","alias_value":"GKKKN2GZP23IE5UK","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_8","alias_value":"GKKKN2GZ","created_at":"2026-05-18T12:32:25.280505+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1906.09821","citing_title":"Classification and Clustering of Arguments with Contextualized Word Embeddings","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"1906.12230","citing_title":"FIESTA: Fast IdEntification of State-of-The-Art models using adaptive bandit algorithms","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"1908.10084","citing_title":"Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4","json":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4.json","graph_json":"https://pith.science/api/pith-number/GKKKN2GZP23IE5UKWRYL3NCGX4/graph.json","events_json":"https://pith.science/api/pith-number/GKKKN2GZP23IE5UKWRYL3NCGX4/events.json","paper":"https://pith.science/paper/GKKKN2GZ"},"agent_actions":{"view_html":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4","download_json":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4.json","view_paper":"https://pith.science/paper/GKKKN2GZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1803.09578&json=true","fetch_graph":"https://pith.science/api/pith-number/GKKKN2GZP23IE5UKWRYL3NCGX4/graph.json","fetch_events":"https://pith.science/api/pith-number/GKKKN2GZP23IE5UKWRYL3NCGX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4/action/storage_attestation","attest_author":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4/action/author_attestation","sign_citation":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4/action/citation_signature","submit_replication":"https://pith.science/pith/GKKKN2GZP23IE5UKWRYL3NCGX4/action/replication_record"}},"created_at":"2026-05-18T00:20:11.056547+00:00","updated_at":"2026-05-18T00:20:11.056547+00:00"}