{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:EZ7OCXRAOZTCZFKQSW47SEJBMX","short_pith_number":"pith:EZ7OCXRA","schema_version":"1.0","canonical_sha256":"267ee15e2076662c955095b9f9112165ea53436b6f6f8db0cb01202b3c7dc829","source":{"kind":"arxiv","id":"2105.14289","version":1},"attestation_state":"computed","paper":{"title":"Modeling Discriminative Representations for Out-of-Domain Detection with Supervised Contrastive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hong Xu, Huixing Jiang, Keqing He, Weiran Xu, Yanan Wu, Yuanmeng Yan, Zhiyuan Zeng, Zijun Liu","submitted_at":"2021-05-29T12:54:22Z","abstract_excerpt":"Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. A key challenge of OOD detection is to learn discriminative semantic features. Traditional cross-entropy loss only focuses on whether a sample is correctly classified, and does not explicitly distinguish the margins between categories. In this paper, we propose a supervised contrastive learning objective to minimize intra-class variance by pulling together in-domain intents belonging to the same class and maximize inter-class variance by pushing apart samples from different classes"},"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":"2105.14289","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-05-29T12:54:22Z","cross_cats_sorted":[],"title_canon_sha256":"48a883b422a0d017c5cf2f2c2d29aee363f888477c293e9007325764abd3efed","abstract_canon_sha256":"5873799ed64723651ee6eb69c16301c1a4a80735ab504d47e2b10cbbd147868a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:44:22.264652Z","signature_b64":"m7/ourG9+Jdyn2SB7kH9MKUkSbBWOABc7kthzukD7rKEJo1ifrTVsGCCOlC672ejCFSZdyly5rzPNYOGA5FsCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"267ee15e2076662c955095b9f9112165ea53436b6f6f8db0cb01202b3c7dc829","last_reissued_at":"2026-07-05T02:44:22.264243Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:44:22.264243Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modeling Discriminative Representations for Out-of-Domain Detection with Supervised Contrastive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hong Xu, Huixing Jiang, Keqing He, Weiran Xu, Yanan Wu, Yuanmeng Yan, Zhiyuan Zeng, Zijun Liu","submitted_at":"2021-05-29T12:54:22Z","abstract_excerpt":"Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. A key challenge of OOD detection is to learn discriminative semantic features. Traditional cross-entropy loss only focuses on whether a sample is correctly classified, and does not explicitly distinguish the margins between categories. In this paper, we propose a supervised contrastive learning objective to minimize intra-class variance by pulling together in-domain intents belonging to the same class and maximize inter-class variance by pushing apart samples from different classes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.14289","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/2105.14289/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":"2105.14289","created_at":"2026-07-05T02:44:22.264302+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.14289v1","created_at":"2026-07-05T02:44:22.264302+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.14289","created_at":"2026-07-05T02:44:22.264302+00:00"},{"alias_kind":"pith_short_12","alias_value":"EZ7OCXRAOZTC","created_at":"2026-07-05T02:44:22.264302+00:00"},{"alias_kind":"pith_short_16","alias_value":"EZ7OCXRAOZTCZFKQ","created_at":"2026-07-05T02:44:22.264302+00:00"},{"alias_kind":"pith_short_8","alias_value":"EZ7OCXRA","created_at":"2026-07-05T02:44:22.264302+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/EZ7OCXRAOZTCZFKQSW47SEJBMX","json":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX.json","graph_json":"https://pith.science/api/pith-number/EZ7OCXRAOZTCZFKQSW47SEJBMX/graph.json","events_json":"https://pith.science/api/pith-number/EZ7OCXRAOZTCZFKQSW47SEJBMX/events.json","paper":"https://pith.science/paper/EZ7OCXRA"},"agent_actions":{"view_html":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX","download_json":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX.json","view_paper":"https://pith.science/paper/EZ7OCXRA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.14289&json=true","fetch_graph":"https://pith.science/api/pith-number/EZ7OCXRAOZTCZFKQSW47SEJBMX/graph.json","fetch_events":"https://pith.science/api/pith-number/EZ7OCXRAOZTCZFKQSW47SEJBMX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX/action/storage_attestation","attest_author":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX/action/author_attestation","sign_citation":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX/action/citation_signature","submit_replication":"https://pith.science/pith/EZ7OCXRAOZTCZFKQSW47SEJBMX/action/replication_record"}},"created_at":"2026-07-05T02:44:22.264302+00:00","updated_at":"2026-07-05T02:44:22.264302+00:00"}