{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TONOWF7INMM7EQMOEFJOPZRC6K","short_pith_number":"pith:TONOWF7I","schema_version":"1.0","canonical_sha256":"9b9aeb17e86b19f2418e2152e7e622f289ee2b49a48c891ddf6f184542161157","source":{"kind":"arxiv","id":"2110.05730","version":2},"attestation_state":"computed","paper":{"title":"Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Hongzhi Yin, Ruihong Qiu, Zi Huang, Zijian Wang","submitted_at":"2021-10-12T04:22:55Z","abstract_excerpt":"Recent advancements of sequential deep learning models such as Transformer and BERT have significantly facilitated the sequential recommendation. However, according to our study, the distribution of item embeddings generated by these models tends to degenerate into an anisotropic shape, which may result in high semantic similarities among embeddings. In this paper, both empirical and theoretical investigations of this representation degeneration problem are first provided, based on which a novel recommender model DuoRec is proposed to improve the item embeddings distribution. Specifically, in "},"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":"2110.05730","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2021-10-12T04:22:55Z","cross_cats_sorted":[],"title_canon_sha256":"b9145a33841d73b3ebcbe0f6b79de90fc069decf92c718d5ed3b1124135c46d0","abstract_canon_sha256":"34408660041e95548910c697620db361fb2a455fbbe6e1b73a39f136b96455f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:33:08.319288Z","signature_b64":"uDKR0dnQSsn25M+EEnTuwMpxfy+007q9uwTAVR8sP6MPMFUeHmHS3kfuuWVz/uuHqslKPvLh2maOVtyHMdh8CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b9aeb17e86b19f2418e2152e7e622f289ee2b49a48c891ddf6f184542161157","last_reissued_at":"2026-07-05T03:33:08.318668Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:33:08.318668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Hongzhi Yin, Ruihong Qiu, Zi Huang, Zijian Wang","submitted_at":"2021-10-12T04:22:55Z","abstract_excerpt":"Recent advancements of sequential deep learning models such as Transformer and BERT have significantly facilitated the sequential recommendation. However, according to our study, the distribution of item embeddings generated by these models tends to degenerate into an anisotropic shape, which may result in high semantic similarities among embeddings. In this paper, both empirical and theoretical investigations of this representation degeneration problem are first provided, based on which a novel recommender model DuoRec is proposed to improve the item embeddings distribution. Specifically, in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.05730","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/2110.05730/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":"2110.05730","created_at":"2026-07-05T03:33:08.318739+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.05730v2","created_at":"2026-07-05T03:33:08.318739+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.05730","created_at":"2026-07-05T03:33:08.318739+00:00"},{"alias_kind":"pith_short_12","alias_value":"TONOWF7INMM7","created_at":"2026-07-05T03:33:08.318739+00:00"},{"alias_kind":"pith_short_16","alias_value":"TONOWF7INMM7EQMO","created_at":"2026-07-05T03:33:08.318739+00:00"},{"alias_kind":"pith_short_8","alias_value":"TONOWF7I","created_at":"2026-07-05T03:33:08.318739+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.19178","citing_title":"Relative Contrastive Learning for Sequential Recommendation with Similarity-based Positive Pair Selection","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K","json":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K.json","graph_json":"https://pith.science/api/pith-number/TONOWF7INMM7EQMOEFJOPZRC6K/graph.json","events_json":"https://pith.science/api/pith-number/TONOWF7INMM7EQMOEFJOPZRC6K/events.json","paper":"https://pith.science/paper/TONOWF7I"},"agent_actions":{"view_html":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K","download_json":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K.json","view_paper":"https://pith.science/paper/TONOWF7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.05730&json=true","fetch_graph":"https://pith.science/api/pith-number/TONOWF7INMM7EQMOEFJOPZRC6K/graph.json","fetch_events":"https://pith.science/api/pith-number/TONOWF7INMM7EQMOEFJOPZRC6K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K/action/storage_attestation","attest_author":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K/action/author_attestation","sign_citation":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K/action/citation_signature","submit_replication":"https://pith.science/pith/TONOWF7INMM7EQMOEFJOPZRC6K/action/replication_record"}},"created_at":"2026-07-05T03:33:08.318739+00:00","updated_at":"2026-07-05T03:33:08.318739+00:00"}