{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6J6D2H4GJTU7ZFWR5WOQ46B7RZ","short_pith_number":"pith:6J6D2H4G","schema_version":"1.0","canonical_sha256":"f27c3d1f864ce9fc96d1ed9d0e783f8e6b06cd71cd2d492b84f3278bf0b47980","source":{"kind":"arxiv","id":"2412.18219","version":2},"attestation_state":"computed","paper":{"title":"Adapter Merging with Centroid Prototype Mapping for Scalable Class-Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hiroshi Kera, Kazuhiko Kawamoto, Takuma Fukuda","submitted_at":"2024-12-24T06:57:16Z","abstract_excerpt":"We propose Adapter Merging with Centroid Prototype Mapping (ACMap), an exemplar-free framework for class-incremental learning (CIL) that addresses both catastrophic forgetting and scalability. While existing methods involve a trade-off between inference time and accuracy, ACMap consolidates task-specific adapters into a single adapter, thus achieving constant inference time across tasks without sacrificing accuracy. The framework employs adapter merging to build a shared subspace that aligns task representations and mitigates forgetting, while centroid prototype mapping maintains high accuracy"},"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":"2412.18219","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-24T06:57:16Z","cross_cats_sorted":[],"title_canon_sha256":"cd5e18b3e3d3c7f9f094b9c8b08dbff303f091bfd3a2abc3f44898657adc556c","abstract_canon_sha256":"dcdd2140deced500408068973721c8726e269e12486296286873891480f40a82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:37:41.226606Z","signature_b64":"jRSWLokIL98kP1HgZiA74lm31hESnXUh7jytVeKt1Qjf+3X76eY/xHKWLPfFskNKkdLkkFYFFO0EVJSjDzt+Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f27c3d1f864ce9fc96d1ed9d0e783f8e6b06cd71cd2d492b84f3278bf0b47980","last_reissued_at":"2026-07-05T10:37:41.226027Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:37:41.226027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapter Merging with Centroid Prototype Mapping for Scalable Class-Incremental Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hiroshi Kera, Kazuhiko Kawamoto, Takuma Fukuda","submitted_at":"2024-12-24T06:57:16Z","abstract_excerpt":"We propose Adapter Merging with Centroid Prototype Mapping (ACMap), an exemplar-free framework for class-incremental learning (CIL) that addresses both catastrophic forgetting and scalability. While existing methods involve a trade-off between inference time and accuracy, ACMap consolidates task-specific adapters into a single adapter, thus achieving constant inference time across tasks without sacrificing accuracy. The framework employs adapter merging to build a shared subspace that aligns task representations and mitigates forgetting, while centroid prototype mapping maintains high accuracy"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.18219","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/2412.18219/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":"2412.18219","created_at":"2026-07-05T10:37:41.226096+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.18219v2","created_at":"2026-07-05T10:37:41.226096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.18219","created_at":"2026-07-05T10:37:41.226096+00:00"},{"alias_kind":"pith_short_12","alias_value":"6J6D2H4GJTU7","created_at":"2026-07-05T10:37:41.226096+00:00"},{"alias_kind":"pith_short_16","alias_value":"6J6D2H4GJTU7ZFWR","created_at":"2026-07-05T10:37:41.226096+00:00"},{"alias_kind":"pith_short_8","alias_value":"6J6D2H4G","created_at":"2026-07-05T10:37:41.226096+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30190","citing_title":"Few-Shot Domain Incremental Learning via Continual Vision-Language Consolidation","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ","json":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ.json","graph_json":"https://pith.science/api/pith-number/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/graph.json","events_json":"https://pith.science/api/pith-number/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/events.json","paper":"https://pith.science/paper/6J6D2H4G"},"agent_actions":{"view_html":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ","download_json":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ.json","view_paper":"https://pith.science/paper/6J6D2H4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.18219&json=true","fetch_graph":"https://pith.science/api/pith-number/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/graph.json","fetch_events":"https://pith.science/api/pith-number/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/action/storage_attestation","attest_author":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/action/author_attestation","sign_citation":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/action/citation_signature","submit_replication":"https://pith.science/pith/6J6D2H4GJTU7ZFWR5WOQ46B7RZ/action/replication_record"}},"created_at":"2026-07-05T10:37:41.226096+00:00","updated_at":"2026-07-05T10:37:41.226096+00:00"}