{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GMCBD6YA3PJZGEP2BHBQE3NHHH","short_pith_number":"pith:GMCBD6YA","schema_version":"1.0","canonical_sha256":"330411fb00dbd39311fa09c3026da739f20dc956ee3fe403d834abbdb21ec1ee","source":{"kind":"arxiv","id":"2106.11930","version":4},"attestation_state":"computed","paper":{"title":"On the importance of cross-task features for class-incremental learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Albin Soutif--Cormerais, Bart{\\l}omiej Twardowski, Joost Van De Weijer, Marc Masana","submitted_at":"2021-06-22T17:03:15Z","abstract_excerpt":"In class-incremental learning, an agent with limited resources needs to learn a sequence of classification tasks, forming an ever growing classification problem, with the constraint of not being able to access data from previous tasks. The main difference with task-incremental learning, where a task-ID is available at inference time, is that the learner also needs to perform cross-task discrimination, i.e. distinguish between classes that have not been seen together. Approaches to tackle this problem are numerous and mostly make use of an external memory (buffer) of non-negligible size. In thi"},"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":"2106.11930","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-22T17:03:15Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"dcdeb6102b31b5316bb0478b45f93871faf5958453992e2c180e7185436b3b88","abstract_canon_sha256":"de48033f850b1aed51818d61fbdd47eb7170d136818cfbabb3bfd58ad0bd6bf1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:50.821404Z","signature_b64":"EpyeVaULbsXtkVciUYCpQnxBP8p3D+aKKFbAhWsyD5OmP3LCSthH+r0A4N0bL7AqSOS/0pwaPy2ZZsgXxJ0WAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"330411fb00dbd39311fa09c3026da739f20dc956ee3fe403d834abbdb21ec1ee","last_reissued_at":"2026-07-05T08:23:50.820974Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:50.820974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the importance of cross-task features for class-incremental learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Albin Soutif--Cormerais, Bart{\\l}omiej Twardowski, Joost Van De Weijer, Marc Masana","submitted_at":"2021-06-22T17:03:15Z","abstract_excerpt":"In class-incremental learning, an agent with limited resources needs to learn a sequence of classification tasks, forming an ever growing classification problem, with the constraint of not being able to access data from previous tasks. The main difference with task-incremental learning, where a task-ID is available at inference time, is that the learner also needs to perform cross-task discrimination, i.e. distinguish between classes that have not been seen together. Approaches to tackle this problem are numerous and mostly make use of an external memory (buffer) of non-negligible size. In thi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.11930","kind":"arxiv","version":4},"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/2106.11930/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":"2106.11930","created_at":"2026-07-05T08:23:50.821031+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.11930v4","created_at":"2026-07-05T08:23:50.821031+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.11930","created_at":"2026-07-05T08:23:50.821031+00:00"},{"alias_kind":"pith_short_12","alias_value":"GMCBD6YA3PJZ","created_at":"2026-07-05T08:23:50.821031+00:00"},{"alias_kind":"pith_short_16","alias_value":"GMCBD6YA3PJZGEP2","created_at":"2026-07-05T08:23:50.821031+00:00"},{"alias_kind":"pith_short_8","alias_value":"GMCBD6YA","created_at":"2026-07-05T08:23:50.821031+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.11673","citing_title":"Contrastive Regularization over LoRA for Multimodal Biomedical Image Incremental Learning","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH","json":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH.json","graph_json":"https://pith.science/api/pith-number/GMCBD6YA3PJZGEP2BHBQE3NHHH/graph.json","events_json":"https://pith.science/api/pith-number/GMCBD6YA3PJZGEP2BHBQE3NHHH/events.json","paper":"https://pith.science/paper/GMCBD6YA"},"agent_actions":{"view_html":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH","download_json":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH.json","view_paper":"https://pith.science/paper/GMCBD6YA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.11930&json=true","fetch_graph":"https://pith.science/api/pith-number/GMCBD6YA3PJZGEP2BHBQE3NHHH/graph.json","fetch_events":"https://pith.science/api/pith-number/GMCBD6YA3PJZGEP2BHBQE3NHHH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH/action/storage_attestation","attest_author":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH/action/author_attestation","sign_citation":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH/action/citation_signature","submit_replication":"https://pith.science/pith/GMCBD6YA3PJZGEP2BHBQE3NHHH/action/replication_record"}},"created_at":"2026-07-05T08:23:50.821031+00:00","updated_at":"2026-07-05T08:23:50.821031+00:00"}