{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:AEEJ2GYBV25F6H3ZXSOMTX6AA3","short_pith_number":"pith:AEEJ2GYB","schema_version":"1.0","canonical_sha256":"01089d1b01aeba5f1f79bc9cc9dfc006cbfeea7ad8caf0abacff8a8e5b5beec2","source":{"kind":"arxiv","id":"2205.13346","version":2},"attestation_state":"computed","paper":{"title":"Keywords and Instances: A Hierarchical Contrastive Learning Framework Unifying Hybrid Granularities for Text Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dongyan Zhao, Feng Wang, Jinxiong Chang, Mingzhe Li, Qishen Zhang, Rui Yan, Taifeng Wang, Wei Chu, Xiexiong Lin, Xiuying Chen, Zhongyi Liu","submitted_at":"2022-05-26T13:26:03Z","abstract_excerpt":"Contrastive learning has achieved impressive success in generation tasks to militate the \"exposure bias\" problem and discriminatively exploit the different quality of references. Existing works mostly focus on contrastive learning on the instance-level without discriminating the contribution of each word, while keywords are the gist of the text and dominant the constrained mapping relationships. Hence, in this work, we propose a hierarchical contrastive learning mechanism, which can unify hybrid granularities semantic meaning in the input text. Concretely, we first propose a keyword graph via "},"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":"2205.13346","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-05-26T13:26:03Z","cross_cats_sorted":[],"title_canon_sha256":"f806e81439943542198024dade29cc399f8e5ad4758b7801b54c6640b19abf77","abstract_canon_sha256":"d07be65c7c73cf41cab04ca38811d9d91f73784551ef23900b6a019305cc0c7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:09.540434Z","signature_b64":"cZAYxU8T6p5LuZW7cyb/z7NkPdnrvQthTBmNncmsul4mYcJqnzaKP/1xVyG4/0XprLHpZ6/uAUaJGFNEkPbMAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01089d1b01aeba5f1f79bc9cc9dfc006cbfeea7ad8caf0abacff8a8e5b5beec2","last_reissued_at":"2026-07-05T10:28:09.539801Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:09.539801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Keywords and Instances: A Hierarchical Contrastive Learning Framework Unifying Hybrid Granularities for Text Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dongyan Zhao, Feng Wang, Jinxiong Chang, Mingzhe Li, Qishen Zhang, Rui Yan, Taifeng Wang, Wei Chu, Xiexiong Lin, Xiuying Chen, Zhongyi Liu","submitted_at":"2022-05-26T13:26:03Z","abstract_excerpt":"Contrastive learning has achieved impressive success in generation tasks to militate the \"exposure bias\" problem and discriminatively exploit the different quality of references. Existing works mostly focus on contrastive learning on the instance-level without discriminating the contribution of each word, while keywords are the gist of the text and dominant the constrained mapping relationships. Hence, in this work, we propose a hierarchical contrastive learning mechanism, which can unify hybrid granularities semantic meaning in the input text. Concretely, we first propose a keyword graph via "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.13346","kind":"arxiv","version":2},"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/2205.13346/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":"2205.13346","created_at":"2026-07-05T10:28:09.539875+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.13346v2","created_at":"2026-07-05T10:28:09.539875+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.13346","created_at":"2026-07-05T10:28:09.539875+00:00"},{"alias_kind":"pith_short_12","alias_value":"AEEJ2GYBV25F","created_at":"2026-07-05T10:28:09.539875+00:00"},{"alias_kind":"pith_short_16","alias_value":"AEEJ2GYBV25F6H3Z","created_at":"2026-07-05T10:28:09.539875+00:00"},{"alias_kind":"pith_short_8","alias_value":"AEEJ2GYB","created_at":"2026-07-05T10:28:09.539875+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/AEEJ2GYBV25F6H3ZXSOMTX6AA3","json":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3.json","graph_json":"https://pith.science/api/pith-number/AEEJ2GYBV25F6H3ZXSOMTX6AA3/graph.json","events_json":"https://pith.science/api/pith-number/AEEJ2GYBV25F6H3ZXSOMTX6AA3/events.json","paper":"https://pith.science/paper/AEEJ2GYB"},"agent_actions":{"view_html":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3","download_json":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3.json","view_paper":"https://pith.science/paper/AEEJ2GYB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.13346&json=true","fetch_graph":"https://pith.science/api/pith-number/AEEJ2GYBV25F6H3ZXSOMTX6AA3/graph.json","fetch_events":"https://pith.science/api/pith-number/AEEJ2GYBV25F6H3ZXSOMTX6AA3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3/action/storage_attestation","attest_author":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3/action/author_attestation","sign_citation":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3/action/citation_signature","submit_replication":"https://pith.science/pith/AEEJ2GYBV25F6H3ZXSOMTX6AA3/action/replication_record"}},"created_at":"2026-07-05T10:28:09.539875+00:00","updated_at":"2026-07-05T10:28:09.539875+00:00"}