{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OY4NTDJFPJXCLWD7V4FPGCGJAD","short_pith_number":"pith:OY4NTDJF","schema_version":"1.0","canonical_sha256":"7638d98d257a6e25d87faf0af308c900fb9bb56f40f1ee6f58838b253cdec810","source":{"kind":"arxiv","id":"2205.13218","version":2},"attestation_state":"computed","paper":{"title":"A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Da-Wei Zhou, De-Chuan Zhan, Han-Jia Ye, Qi-Wei Wang","submitted_at":"2022-05-26T08:24:01Z","abstract_excerpt":"Real-world applications require the classification model to adapt to new classes without forgetting old ones. Correspondingly, Class-Incremental Learning (CIL) aims to train a model with limited memory size to meet this requirement. Typical CIL methods tend to save representative exemplars from former classes to resist forgetting, while recent works find that storing models from history can substantially boost the performance. However, the stored models are not counted into the memory budget, which implicitly results in unfair comparisons. We find that when counting the model size into the tot"},"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":"2205.13218","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-26T08:24:01Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"c99ff433df9b2743180c4c7479db722ce90564d4814ee28aa5080884a571b0f5","abstract_canon_sha256":"489b8a49708815595fd5d1ee7c00d888917f51425ef89a9ace5d726d238ab112"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:42:25.335483Z","signature_b64":"2K0luZLXY5MoYZgfyWpbmgGOCv2mk4Z53QwfAgTGGsGYhJRTNVJHcQLIhOv9IDIb5vrtbcxAK9cPkmlk3VPhAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7638d98d257a6e25d87faf0af308c900fb9bb56f40f1ee6f58838b253cdec810","last_reissued_at":"2026-07-05T05:42:25.334991Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:42:25.334991Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Da-Wei Zhou, De-Chuan Zhan, Han-Jia Ye, Qi-Wei Wang","submitted_at":"2022-05-26T08:24:01Z","abstract_excerpt":"Real-world applications require the classification model to adapt to new classes without forgetting old ones. Correspondingly, Class-Incremental Learning (CIL) aims to train a model with limited memory size to meet this requirement. Typical CIL methods tend to save representative exemplars from former classes to resist forgetting, while recent works find that storing models from history can substantially boost the performance. However, the stored models are not counted into the memory budget, which implicitly results in unfair comparisons. We find that when counting the model size into the tot"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.13218","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/2205.13218/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":"2205.13218","created_at":"2026-07-05T05:42:25.335044+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.13218v2","created_at":"2026-07-05T05:42:25.335044+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.13218","created_at":"2026-07-05T05:42:25.335044+00:00"},{"alias_kind":"pith_short_12","alias_value":"OY4NTDJFPJXC","created_at":"2026-07-05T05:42:25.335044+00:00"},{"alias_kind":"pith_short_16","alias_value":"OY4NTDJFPJXCLWD7","created_at":"2026-07-05T05:42:25.335044+00:00"},{"alias_kind":"pith_short_8","alias_value":"OY4NTDJF","created_at":"2026-07-05T05:42:25.335044+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09960","citing_title":"HydraCIL: Decoupled Class-Incremental Learning through Prototype-Guided Multi-Head Classifiers","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2411.02813","citing_title":"Sparse Orthogonal Parameters Tuning for Continual Learning","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2506.21035","citing_title":"Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2512.03537","citing_title":"Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02765","citing_title":"Towards Realistic Class-Incremental Learning with Free-Flow Increments","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11742","citing_title":"Online Continual Learning with Dynamic Label Hierarchies","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11904","citing_title":"Beyond Point-wise Neural Collapse: A Topology-Aware Hierarchical Classifier for Class-Incremental Learning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03866","citing_title":"Memory-Efficient Continual Learning with CLIP Models","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD","json":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD.json","graph_json":"https://pith.science/api/pith-number/OY4NTDJFPJXCLWD7V4FPGCGJAD/graph.json","events_json":"https://pith.science/api/pith-number/OY4NTDJFPJXCLWD7V4FPGCGJAD/events.json","paper":"https://pith.science/paper/OY4NTDJF"},"agent_actions":{"view_html":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD","download_json":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD.json","view_paper":"https://pith.science/paper/OY4NTDJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.13218&json=true","fetch_graph":"https://pith.science/api/pith-number/OY4NTDJFPJXCLWD7V4FPGCGJAD/graph.json","fetch_events":"https://pith.science/api/pith-number/OY4NTDJFPJXCLWD7V4FPGCGJAD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD/action/storage_attestation","attest_author":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD/action/author_attestation","sign_citation":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD/action/citation_signature","submit_replication":"https://pith.science/pith/OY4NTDJFPJXCLWD7V4FPGCGJAD/action/replication_record"}},"created_at":"2026-07-05T05:42:25.335044+00:00","updated_at":"2026-07-05T05:42:25.335044+00:00"}