{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:I2D5KEU2MHSY3STPH43CCLNRTU","short_pith_number":"pith:I2D5KEU2","schema_version":"1.0","canonical_sha256":"4687d5129a61e58dca6f3f36212db19d237f07dd664bf3be5330a29f305931bb","source":{"kind":"arxiv","id":"2405.15633","version":1},"attestation_state":"computed","paper":{"title":"Less is more: Summarizing Patch Tokens for efficient Multi-Label Class-Incremental Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Elisa Ricci, Massimiliano Mancini, St\\'ephane Lathuili\\`ere, Subhankar Roy, Thomas De Min","submitted_at":"2024-05-24T15:18:27Z","abstract_excerpt":"Prompt tuning has emerged as an effective rehearsal-free technique for class-incremental learning (CIL) that learns a tiny set of task-specific parameters (or prompts) to instruct a pre-trained transformer to learn on a sequence of tasks. Albeit effective, prompt tuning methods do not lend well in the multi-label class incremental learning (MLCIL) scenario (where an image contains multiple foreground classes) due to the ambiguity in selecting the correct prompt(s) corresponding to different foreground objects belonging to multiple tasks. To circumvent this issue we propose to eliminate the pro"},"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":"2405.15633","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-24T15:18:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3ae97401d3ac750380336ca87eba58c0457bd78507d0e324e58c6a7b2b08c2e9","abstract_canon_sha256":"17c7e6a0d46e35502ccde9fd9400efa35ba3f1072dae10ba0c8677aa77eed62a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:51.894552Z","signature_b64":"cXiDTs78AhT7yTOIM8G2UXK1TtGfKjCwgVRw1KeV3n7DDyQ0sDVquWGDZ/7wbmejUSo45HHuw3O/uTALwL03Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4687d5129a61e58dca6f3f36212db19d237f07dd664bf3be5330a29f305931bb","last_reissued_at":"2026-07-05T08:22:51.894067Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:51.894067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Less is more: Summarizing Patch Tokens for efficient Multi-Label Class-Incremental Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Elisa Ricci, Massimiliano Mancini, St\\'ephane Lathuili\\`ere, Subhankar Roy, Thomas De Min","submitted_at":"2024-05-24T15:18:27Z","abstract_excerpt":"Prompt tuning has emerged as an effective rehearsal-free technique for class-incremental learning (CIL) that learns a tiny set of task-specific parameters (or prompts) to instruct a pre-trained transformer to learn on a sequence of tasks. Albeit effective, prompt tuning methods do not lend well in the multi-label class incremental learning (MLCIL) scenario (where an image contains multiple foreground classes) due to the ambiguity in selecting the correct prompt(s) corresponding to different foreground objects belonging to multiple tasks. To circumvent this issue we propose to eliminate the pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15633","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/2405.15633/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":"2405.15633","created_at":"2026-07-05T08:22:51.894126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.15633v1","created_at":"2026-07-05T08:22:51.894126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15633","created_at":"2026-07-05T08:22:51.894126+00:00"},{"alias_kind":"pith_short_12","alias_value":"I2D5KEU2MHSY","created_at":"2026-07-05T08:22:51.894126+00:00"},{"alias_kind":"pith_short_16","alias_value":"I2D5KEU2MHSY3STP","created_at":"2026-07-05T08:22:51.894126+00:00"},{"alias_kind":"pith_short_8","alias_value":"I2D5KEU2","created_at":"2026-07-05T08:22:51.894126+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/I2D5KEU2MHSY3STPH43CCLNRTU","json":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU.json","graph_json":"https://pith.science/api/pith-number/I2D5KEU2MHSY3STPH43CCLNRTU/graph.json","events_json":"https://pith.science/api/pith-number/I2D5KEU2MHSY3STPH43CCLNRTU/events.json","paper":"https://pith.science/paper/I2D5KEU2"},"agent_actions":{"view_html":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU","download_json":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU.json","view_paper":"https://pith.science/paper/I2D5KEU2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.15633&json=true","fetch_graph":"https://pith.science/api/pith-number/I2D5KEU2MHSY3STPH43CCLNRTU/graph.json","fetch_events":"https://pith.science/api/pith-number/I2D5KEU2MHSY3STPH43CCLNRTU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU/action/storage_attestation","attest_author":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU/action/author_attestation","sign_citation":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU/action/citation_signature","submit_replication":"https://pith.science/pith/I2D5KEU2MHSY3STPH43CCLNRTU/action/replication_record"}},"created_at":"2026-07-05T08:22:51.894126+00:00","updated_at":"2026-07-05T08:22:51.894126+00:00"}