{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6QBUAIRQN5NZMBDIAWV2WSROGE","short_pith_number":"pith:6QBUAIRQ","schema_version":"1.0","canonical_sha256":"f4034022306f5b96046805abab4a2e31377be9e6b6954399e323fae4ea2b60c9","source":{"kind":"arxiv","id":"2504.06780","version":1},"attestation_state":"computed","paper":{"title":"CHIME: A Compressive Framework for Holistic Interest Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Kaiyuan Li, Kun Gai, Peng Jiang, Rui Xiang, Xialong Liu, Yanhua Cheng, Yong Bai, Yongxiang Tang","submitted_at":"2025-04-09T11:08:49Z","abstract_excerpt":"Modeling holistic user interests is important for improving recommendation systems but is challenged by high computational cost and difficulty in handling diverse information with full behavior context. Existing search-based methods might lose critical signals during behavior selection. To overcome these limitations, we propose CHIME: A Compressive Framework for Holistic Interest Modeling. It uses adapted large language models to encode complete user behaviors with heterogeneous inputs. We introduce multi-granular contrastive learning objectives to capture both persistent and transient interes"},"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":"2504.06780","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-09T11:08:49Z","cross_cats_sorted":[],"title_canon_sha256":"d47e22fcefdc6646ba005d7a97c1d4f669bdb28841aef1617f0204b876a239c0","abstract_canon_sha256":"63645fd3f330a963df216a2c61e26d40d59fee1641521341476fb433d6c5484c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:45.821538Z","signature_b64":"+bDPQS/sTuNR8pW+ccYmVHilO7OU9sJAEPjGrnySuDMngslSxLJmSHeDQTRBS18DTa9ebtrpjx4BIzF/T6guAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4034022306f5b96046805abab4a2e31377be9e6b6954399e323fae4ea2b60c9","last_reissued_at":"2026-07-05T10:46:45.820987Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:45.820987Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CHIME: A Compressive Framework for Holistic Interest Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Kaiyuan Li, Kun Gai, Peng Jiang, Rui Xiang, Xialong Liu, Yanhua Cheng, Yong Bai, Yongxiang Tang","submitted_at":"2025-04-09T11:08:49Z","abstract_excerpt":"Modeling holistic user interests is important for improving recommendation systems but is challenged by high computational cost and difficulty in handling diverse information with full behavior context. Existing search-based methods might lose critical signals during behavior selection. To overcome these limitations, we propose CHIME: A Compressive Framework for Holistic Interest Modeling. It uses adapted large language models to encode complete user behaviors with heterogeneous inputs. We introduce multi-granular contrastive learning objectives to capture both persistent and transient interes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06780","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/2504.06780/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":"2504.06780","created_at":"2026-07-05T10:46:45.821046+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.06780v1","created_at":"2026-07-05T10:46:45.821046+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06780","created_at":"2026-07-05T10:46:45.821046+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QBUAIRQN5NZ","created_at":"2026-07-05T10:46:45.821046+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QBUAIRQN5NZMBDI","created_at":"2026-07-05T10:46:45.821046+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QBUAIRQ","created_at":"2026-07-05T10:46:45.821046+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03692","citing_title":"SITA: Semantic Interest Tokens for Target-Aware Compression in Long-Sequence Recommendation","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE","json":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE.json","graph_json":"https://pith.science/api/pith-number/6QBUAIRQN5NZMBDIAWV2WSROGE/graph.json","events_json":"https://pith.science/api/pith-number/6QBUAIRQN5NZMBDIAWV2WSROGE/events.json","paper":"https://pith.science/paper/6QBUAIRQ"},"agent_actions":{"view_html":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE","download_json":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE.json","view_paper":"https://pith.science/paper/6QBUAIRQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.06780&json=true","fetch_graph":"https://pith.science/api/pith-number/6QBUAIRQN5NZMBDIAWV2WSROGE/graph.json","fetch_events":"https://pith.science/api/pith-number/6QBUAIRQN5NZMBDIAWV2WSROGE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE/action/storage_attestation","attest_author":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE/action/author_attestation","sign_citation":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE/action/citation_signature","submit_replication":"https://pith.science/pith/6QBUAIRQN5NZMBDIAWV2WSROGE/action/replication_record"}},"created_at":"2026-07-05T10:46:45.821046+00:00","updated_at":"2026-07-05T10:46:45.821046+00:00"}