{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:X6PUXE647DI6PY7YNZIQZ7YPMN","short_pith_number":"pith:X6PUXE64","schema_version":"1.0","canonical_sha256":"bf9f4b93dcf8d1e7e3f86e510cff0f637a99c5266533fadb714dd80def9da151","source":{"kind":"arxiv","id":"2307.02078","version":1},"attestation_state":"computed","paper":{"title":"Graph Contrastive Topic Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Lei Liu, Qianqian Xie, Sophia Ananiadou, Zheheng Luo","submitted_at":"2023-07-05T07:39:47Z","abstract_excerpt":"Existing NTMs with contrastive learning suffer from the sample bias problem owing to the word frequency-based sampling strategy, which may result in false negative samples with similar semantics to the prototypes. In this paper, we aim to explore the efficient sampling strategy and contrastive learning in NTMs to address the aforementioned issue. We propose a new sampling assumption that negative samples should contain words that are semantically irrelevant to the prototype. Based on it, we propose the graph contrastive topic model (GCTM), which conducts graph contrastive learning (GCL) using "},"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":"2307.02078","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T07:39:47Z","cross_cats_sorted":[],"title_canon_sha256":"b3986d189d7159cfb4b68b0faef1e42170891077906eb94bb7adf416c01bcc67","abstract_canon_sha256":"fb5617f2dba73270cd536b633d5633dc7f1bc1f482769c81997dc97979393b02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:28:05.043882Z","signature_b64":"K1a3Ix8xCWPnez7HKDxYFcG8AbVXv297KXcgYppqJEclEkQSCzlijTL37hX59ajSSZL9tNido+Ow9sPD3BFkDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf9f4b93dcf8d1e7e3f86e510cff0f637a99c5266533fadb714dd80def9da151","last_reissued_at":"2026-07-05T06:28:05.043490Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:28:05.043490Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Graph Contrastive Topic Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Lei Liu, Qianqian Xie, Sophia Ananiadou, Zheheng Luo","submitted_at":"2023-07-05T07:39:47Z","abstract_excerpt":"Existing NTMs with contrastive learning suffer from the sample bias problem owing to the word frequency-based sampling strategy, which may result in false negative samples with similar semantics to the prototypes. In this paper, we aim to explore the efficient sampling strategy and contrastive learning in NTMs to address the aforementioned issue. We propose a new sampling assumption that negative samples should contain words that are semantically irrelevant to the prototype. Based on it, we propose the graph contrastive topic model (GCTM), which conducts graph contrastive learning (GCL) using "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.02078","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/2307.02078/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":"2307.02078","created_at":"2026-07-05T06:28:05.043554+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.02078v1","created_at":"2026-07-05T06:28:05.043554+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.02078","created_at":"2026-07-05T06:28:05.043554+00:00"},{"alias_kind":"pith_short_12","alias_value":"X6PUXE647DI6","created_at":"2026-07-05T06:28:05.043554+00:00"},{"alias_kind":"pith_short_16","alias_value":"X6PUXE647DI6PY7Y","created_at":"2026-07-05T06:28:05.043554+00:00"},{"alias_kind":"pith_short_8","alias_value":"X6PUXE64","created_at":"2026-07-05T06:28:05.043554+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.00592","citing_title":"GeoMoE: Divide-and-Conquer Motion Field Modeling with Mixture-of-Experts for Two-View Geometry","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN","json":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN.json","graph_json":"https://pith.science/api/pith-number/X6PUXE647DI6PY7YNZIQZ7YPMN/graph.json","events_json":"https://pith.science/api/pith-number/X6PUXE647DI6PY7YNZIQZ7YPMN/events.json","paper":"https://pith.science/paper/X6PUXE64"},"agent_actions":{"view_html":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN","download_json":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN.json","view_paper":"https://pith.science/paper/X6PUXE64","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.02078&json=true","fetch_graph":"https://pith.science/api/pith-number/X6PUXE647DI6PY7YNZIQZ7YPMN/graph.json","fetch_events":"https://pith.science/api/pith-number/X6PUXE647DI6PY7YNZIQZ7YPMN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN/action/storage_attestation","attest_author":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN/action/author_attestation","sign_citation":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN/action/citation_signature","submit_replication":"https://pith.science/pith/X6PUXE647DI6PY7YNZIQZ7YPMN/action/replication_record"}},"created_at":"2026-07-05T06:28:05.043554+00:00","updated_at":"2026-07-05T06:28:05.043554+00:00"}