{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PFXKH3AC4CSP7LRFJG4D4ZY2ZM","short_pith_number":"pith:PFXKH3AC","schema_version":"1.0","canonical_sha256":"796ea3ec02e0a4ffae2549b83e671acb058f43adec13dfccd849fde8c311da94","source":{"kind":"arxiv","id":"1908.06941","version":1},"attestation_state":"computed","paper":{"title":"Why So Down? The Role of Negative (and Positive) Pointwise Mutual Information in Distributional Semantics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexandre Salle, Aline Villavicencio","submitted_at":"2019-08-19T17:26:13Z","abstract_excerpt":"In distributional semantics, the pointwise mutual information ($\\mathit{PMI}$) weighting of the cooccurrence matrix performs far better than raw counts. There is, however, an issue with unobserved pair cooccurrences as $\\mathit{PMI}$ goes to negative infinity. This problem is aggravated by unreliable statistics from finite corpora which lead to a large number of such pairs. A common practice is to clip negative $\\mathit{PMI}$ ($\\mathit{\\texttt{-} PMI}$) at $0$, also known as Positive $\\mathit{PMI}$ ($\\mathit{PPMI}$). In this paper, we investigate alternative ways of dealing with $\\mathit{\\text"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1908.06941","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-19T17:26:13Z","cross_cats_sorted":[],"title_canon_sha256":"05285c131dc5858bf243d77791965556766b6bdc2a074f225dc5800b7a98d64d","abstract_canon_sha256":"7f61fee0e487cdb64423df33b48315b37869ceaf71d47410545193ac73685044"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:58:21.399303Z","signature_b64":"Z76gOIdJcEXIX4ELjDqevI9jobXM96oeTPlGqrbvOfNyM2dI3e4kYwBv9YwRv4vBmxqNemqRNLzxEELzRx8MBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"796ea3ec02e0a4ffae2549b83e671acb058f43adec13dfccd849fde8c311da94","last_reissued_at":"2026-07-04T23:58:21.398931Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:58:21.398931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why So Down? The Role of Negative (and Positive) Pointwise Mutual Information in Distributional Semantics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexandre Salle, Aline Villavicencio","submitted_at":"2019-08-19T17:26:13Z","abstract_excerpt":"In distributional semantics, the pointwise mutual information ($\\mathit{PMI}$) weighting of the cooccurrence matrix performs far better than raw counts. There is, however, an issue with unobserved pair cooccurrences as $\\mathit{PMI}$ goes to negative infinity. This problem is aggravated by unreliable statistics from finite corpora which lead to a large number of such pairs. A common practice is to clip negative $\\mathit{PMI}$ ($\\mathit{\\texttt{-} PMI}$) at $0$, also known as Positive $\\mathit{PMI}$ ($\\mathit{PPMI}$). In this paper, we investigate alternative ways of dealing with $\\mathit{\\text"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.06941","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/1908.06941/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":"1908.06941","created_at":"2026-07-04T23:58:21.398993+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.06941v1","created_at":"2026-07-04T23:58:21.398993+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.06941","created_at":"2026-07-04T23:58:21.398993+00:00"},{"alias_kind":"pith_short_12","alias_value":"PFXKH3AC4CSP","created_at":"2026-07-04T23:58:21.398993+00:00"},{"alias_kind":"pith_short_16","alias_value":"PFXKH3AC4CSP7LRF","created_at":"2026-07-04T23:58:21.398993+00:00"},{"alias_kind":"pith_short_8","alias_value":"PFXKH3AC","created_at":"2026-07-04T23:58:21.398993+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/PFXKH3AC4CSP7LRFJG4D4ZY2ZM","json":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM.json","graph_json":"https://pith.science/api/pith-number/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/graph.json","events_json":"https://pith.science/api/pith-number/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/events.json","paper":"https://pith.science/paper/PFXKH3AC"},"agent_actions":{"view_html":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM","download_json":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM.json","view_paper":"https://pith.science/paper/PFXKH3AC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.06941&json=true","fetch_graph":"https://pith.science/api/pith-number/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/graph.json","fetch_events":"https://pith.science/api/pith-number/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/action/storage_attestation","attest_author":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/action/author_attestation","sign_citation":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/action/citation_signature","submit_replication":"https://pith.science/pith/PFXKH3AC4CSP7LRFJG4D4ZY2ZM/action/replication_record"}},"created_at":"2026-07-04T23:58:21.398993+00:00","updated_at":"2026-07-04T23:58:21.398993+00:00"}