{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:OOOJM5MVFJCIRAZMR3CUQBGG2Z","short_pith_number":"pith:OOOJM5MV","schema_version":"1.0","canonical_sha256":"739c9675952a4488832c8ec54804c6d67494ea1ef133742903aa933eb59cf62e","source":{"kind":"arxiv","id":"2103.16788","version":1},"attestation_state":"computed","paper":{"title":"DER: Dynamically Expandable Representation for Class Incremental Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiangwei Xie, Shipeng Yan, Xuming He","submitted_at":"2021-03-31T03:16:44Z","abstract_excerpt":"We address the problem of class incremental learning, which is a core step towards achieving adaptive vision intelligence. In particular, we consider the task setting of incremental learning with limited memory and aim to achieve better stability-plasticity trade-off. To this end, we propose a novel two-stage learning approach that utilizes a dynamically expandable representation for more effective incremental concept modeling. Specifically, at each incremental step, we freeze the previously learned representation and augment it with additional feature dimensions from a new learnable feature e"},"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":"2103.16788","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-03-31T03:16:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eaebfb6fb80d03bc804dde34994dc5d4409d15205e161162e19ea4d717fd10e9","abstract_canon_sha256":"19f07772ed04c9412448aa3b7d5ad49b3c661494e415920880be00334d3871c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:27:58.394947Z","signature_b64":"EPikGBO/4bJ6t/aB1r1eSNUDWu1rx8ReziMij3VqLOtsez2nbR1MNDTKpaBKLoGSuAAQ+EtwO1YMzWMyUWPrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"739c9675952a4488832c8ec54804c6d67494ea1ef133742903aa933eb59cf62e","last_reissued_at":"2026-07-05T02:27:58.394422Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:27:58.394422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DER: Dynamically Expandable Representation for Class Incremental Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiangwei Xie, Shipeng Yan, Xuming He","submitted_at":"2021-03-31T03:16:44Z","abstract_excerpt":"We address the problem of class incremental learning, which is a core step towards achieving adaptive vision intelligence. In particular, we consider the task setting of incremental learning with limited memory and aim to achieve better stability-plasticity trade-off. To this end, we propose a novel two-stage learning approach that utilizes a dynamically expandable representation for more effective incremental concept modeling. Specifically, at each incremental step, we freeze the previously learned representation and augment it with additional feature dimensions from a new learnable feature e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.16788","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/2103.16788/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":"2103.16788","created_at":"2026-07-05T02:27:58.394482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.16788v1","created_at":"2026-07-05T02:27:58.394482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.16788","created_at":"2026-07-05T02:27:58.394482+00:00"},{"alias_kind":"pith_short_12","alias_value":"OOOJM5MVFJCI","created_at":"2026-07-05T02:27:58.394482+00:00"},{"alias_kind":"pith_short_16","alias_value":"OOOJM5MVFJCIRAZM","created_at":"2026-07-05T02:27:58.394482+00:00"},{"alias_kind":"pith_short_8","alias_value":"OOOJM5MV","created_at":"2026-07-05T02:27:58.394482+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26097","citing_title":"Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z","json":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z.json","graph_json":"https://pith.science/api/pith-number/OOOJM5MVFJCIRAZMR3CUQBGG2Z/graph.json","events_json":"https://pith.science/api/pith-number/OOOJM5MVFJCIRAZMR3CUQBGG2Z/events.json","paper":"https://pith.science/paper/OOOJM5MV"},"agent_actions":{"view_html":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z","download_json":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z.json","view_paper":"https://pith.science/paper/OOOJM5MV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.16788&json=true","fetch_graph":"https://pith.science/api/pith-number/OOOJM5MVFJCIRAZMR3CUQBGG2Z/graph.json","fetch_events":"https://pith.science/api/pith-number/OOOJM5MVFJCIRAZMR3CUQBGG2Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z/action/storage_attestation","attest_author":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z/action/author_attestation","sign_citation":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z/action/citation_signature","submit_replication":"https://pith.science/pith/OOOJM5MVFJCIRAZMR3CUQBGG2Z/action/replication_record"}},"created_at":"2026-07-05T02:27:58.394482+00:00","updated_at":"2026-07-05T02:27:58.394482+00:00"}