{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XZF67SJJ25LVGT35XF5M4ML4V6","short_pith_number":"pith:XZF67SJJ","schema_version":"1.0","canonical_sha256":"be4befc929d757534f7db97ace317caf9427e4021c962f63791673f4709bde1f","source":{"kind":"arxiv","id":"2305.15756","version":1},"attestation_state":"computed","paper":{"title":"UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Huimin Wang, Kam-Fai Wong, Yiming Du, Zhiming Mao","submitted_at":"2023-05-25T06:11:31Z","abstract_excerpt":"Prior study has shown that pretrained language models (PLM) can boost the performance of text-based recommendation. In contrast to previous works that either use PLM to encode user history as a whole input text, or impose an additional aggregation network to fuse multi-turn history representations, we propose a unified local- and global-attention Transformer encoder to better model two-level contexts of user history. Moreover, conditioned on user history encoded by Transformer encoders, our framework leverages Transformer decoders to estimate the language perplexity of candidate text items, wh"},"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":"2305.15756","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-25T06:11:31Z","cross_cats_sorted":[],"title_canon_sha256":"3d118b731b2124a48915ebcf0c92079b3e59897ba85783c0d32d4b0cbbdcb937","abstract_canon_sha256":"7435a03b4925a28e4d788011991a74719b03ab3233ea609d7d42af57f7974a16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:44.335931Z","signature_b64":"RWSumYp/VbQjAXBXK6UMo4n/czLC89dGGANPIyJJ49R/odwHwnZsWtoBWBmd2/zdK434FSpcB4AuCO7b1U5dAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be4befc929d757534f7db97ace317caf9427e4021c962f63791673f4709bde1f","last_reissued_at":"2026-07-05T06:13:44.335480Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:44.335480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Huimin Wang, Kam-Fai Wong, Yiming Du, Zhiming Mao","submitted_at":"2023-05-25T06:11:31Z","abstract_excerpt":"Prior study has shown that pretrained language models (PLM) can boost the performance of text-based recommendation. In contrast to previous works that either use PLM to encode user history as a whole input text, or impose an additional aggregation network to fuse multi-turn history representations, we propose a unified local- and global-attention Transformer encoder to better model two-level contexts of user history. Moreover, conditioned on user history encoded by Transformer encoders, our framework leverages Transformer decoders to estimate the language perplexity of candidate text items, wh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.15756","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/2305.15756/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":"2305.15756","created_at":"2026-07-05T06:13:44.335544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.15756v1","created_at":"2026-07-05T06:13:44.335544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.15756","created_at":"2026-07-05T06:13:44.335544+00:00"},{"alias_kind":"pith_short_12","alias_value":"XZF67SJJ25LV","created_at":"2026-07-05T06:13:44.335544+00:00"},{"alias_kind":"pith_short_16","alias_value":"XZF67SJJ25LVGT35","created_at":"2026-07-05T06:13:44.335544+00:00"},{"alias_kind":"pith_short_8","alias_value":"XZF67SJJ","created_at":"2026-07-05T06:13:44.335544+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.13344","citing_title":"Full-Stack Optimized Large Language Models for Lifelong Sequential Behavior Comprehension in Recommendation","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6","json":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6.json","graph_json":"https://pith.science/api/pith-number/XZF67SJJ25LVGT35XF5M4ML4V6/graph.json","events_json":"https://pith.science/api/pith-number/XZF67SJJ25LVGT35XF5M4ML4V6/events.json","paper":"https://pith.science/paper/XZF67SJJ"},"agent_actions":{"view_html":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6","download_json":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6.json","view_paper":"https://pith.science/paper/XZF67SJJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.15756&json=true","fetch_graph":"https://pith.science/api/pith-number/XZF67SJJ25LVGT35XF5M4ML4V6/graph.json","fetch_events":"https://pith.science/api/pith-number/XZF67SJJ25LVGT35XF5M4ML4V6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6/action/storage_attestation","attest_author":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6/action/author_attestation","sign_citation":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6/action/citation_signature","submit_replication":"https://pith.science/pith/XZF67SJJ25LVGT35XF5M4ML4V6/action/replication_record"}},"created_at":"2026-07-05T06:13:44.335544+00:00","updated_at":"2026-07-05T06:13:44.335544+00:00"}