{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2014:L7L75S7IAN5HQBDZ36DU7NHVCU","short_pith_number":"pith:L7L75S7I","schema_version":"1.0","canonical_sha256":"5fd7fecbe8037a780479df874fb4f51502c3d925e0472e2122a30aa5928e072d","source":{"kind":"arxiv","id":"1412.6568","version":3},"attestation_state":"computed","paper":{"title":"Improving zero-shot learning by mitigating the hubness problem","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Angeliki Lazaridou, Georgiana Dinu, Marco Baroni","submitted_at":"2014-12-20T01:03:46Z","abstract_excerpt":"The zero-shot paradigm exploits vector-based word representations extracted from text corpora with unsupervised methods to learn general mapping functions from other feature spaces onto word space, where the words associated to the nearest neighbours of the mapped vectors are used as their linguistic labels. We show that the neighbourhoods of the mapped elements are strongly polluted by hubs, vectors that tend to be near a high proportion of items, pushing their correct labels down the neighbour list. After illustrating the problem empirically, we propose a simple method to correct it by takin"},"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":"1412.6568","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2014-12-20T01:03:46Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2fa4f1214b638211b3f60d277cf1bfefa8e26ed98bb42253e475749e15c66d10","abstract_canon_sha256":"5d1087a1263396c81a932494ece5071dc91afc9906f780ecc1763f88f5f7b39b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:18:43.092663Z","signature_b64":"mq4/UczKQn4vaUePemoUl+796mxsFimb9U8eUmGwIcZYiHeVQPVc8LXMzrg+k7bHpl9tGn7OrRYxPGkPa6wkAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fd7fecbe8037a780479df874fb4f51502c3d925e0472e2122a30aa5928e072d","last_reissued_at":"2026-05-18T02:18:43.091953Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:18:43.091953Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving zero-shot learning by mitigating the hubness problem","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Angeliki Lazaridou, Georgiana Dinu, Marco Baroni","submitted_at":"2014-12-20T01:03:46Z","abstract_excerpt":"The zero-shot paradigm exploits vector-based word representations extracted from text corpora with unsupervised methods to learn general mapping functions from other feature spaces onto word space, where the words associated to the nearest neighbours of the mapped vectors are used as their linguistic labels. We show that the neighbourhoods of the mapped elements are strongly polluted by hubs, vectors that tend to be near a high proportion of items, pushing their correct labels down the neighbour list. After illustrating the problem empirically, we propose a simple method to correct it by takin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1412.6568","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":""},"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":"1412.6568","created_at":"2026-05-18T02:18:43.092056+00:00"},{"alias_kind":"arxiv_version","alias_value":"1412.6568v3","created_at":"2026-05-18T02:18:43.092056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1412.6568","created_at":"2026-05-18T02:18:43.092056+00:00"},{"alias_kind":"pith_short_12","alias_value":"L7L75S7IAN5H","created_at":"2026-05-18T12:28:35.611951+00:00"},{"alias_kind":"pith_short_16","alias_value":"L7L75S7IAN5HQBDZ","created_at":"2026-05-18T12:28:35.611951+00:00"},{"alias_kind":"pith_short_8","alias_value":"L7L75S7I","created_at":"2026-05-18T12:28:35.611951+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"1907.00544","citing_title":"Unsupervised Adversarial Graph Alignment with Graph Embedding","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2412.14113","citing_title":"Adversarial Hubness in Multi-Modal Retrieval","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2603.20738","citing_title":"SATTC: Structure-Aware Label-Free Test-Time Calibration for Cross-Subject EEG-to-Image Retrieval","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27674","citing_title":"One Single Hub Text Breaks CLIP: Identifying Vulnerabilities in Cross-Modal Encoders via Hubness","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU","json":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU.json","graph_json":"https://pith.science/api/pith-number/L7L75S7IAN5HQBDZ36DU7NHVCU/graph.json","events_json":"https://pith.science/api/pith-number/L7L75S7IAN5HQBDZ36DU7NHVCU/events.json","paper":"https://pith.science/paper/L7L75S7I"},"agent_actions":{"view_html":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU","download_json":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU.json","view_paper":"https://pith.science/paper/L7L75S7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1412.6568&json=true","fetch_graph":"https://pith.science/api/pith-number/L7L75S7IAN5HQBDZ36DU7NHVCU/graph.json","fetch_events":"https://pith.science/api/pith-number/L7L75S7IAN5HQBDZ36DU7NHVCU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU/action/storage_attestation","attest_author":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU/action/author_attestation","sign_citation":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU/action/citation_signature","submit_replication":"https://pith.science/pith/L7L75S7IAN5HQBDZ36DU7NHVCU/action/replication_record"}},"created_at":"2026-05-18T02:18:43.092056+00:00","updated_at":"2026-05-18T02:18:43.092056+00:00"}