{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:X3ZVCTH6ISBK4W3532EEBYTZAM","short_pith_number":"pith:X3ZVCTH6","schema_version":"1.0","canonical_sha256":"bef3514cfe4482ae5b7dde8840e279031f9d0f7f7978f7fee43e36c8f5814f57","source":{"kind":"arxiv","id":"2312.08888","version":3},"attestation_state":"computed","paper":{"title":"Read Between the Layers: Leveraging Multi-Layer Representations for Rehearsal-Free Continual Learning with Pre-Trained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Hans Hergen Lehmann, Jae Hee Lee, Kyra Ahrens, Stefan Wermter","submitted_at":"2023-12-13T13:11:44Z","abstract_excerpt":"We address the Continual Learning (CL) problem, wherein a model must learn a sequence of tasks from non-stationary distributions while preserving prior knowledge upon encountering new experiences. With the advancement of foundation models, CL research has pivoted from the initial learning-from-scratch paradigm towards utilizing generic features from large-scale pre-training. However, existing approaches to CL with pre-trained models primarily focus on separating class-specific features from the final representation layer and neglect the potential of intermediate representations to capture low-"},"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":"2312.08888","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-13T13:11:44Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"5525f77eb3ed088468b45293454899c0a3b6b5b1773c064967919a80350c3c8a","abstract_canon_sha256":"136d05a68e9f129452699f7b82e9ba5d9465ff51aeecfcfdd5173108f4a85695"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:40:11.266411Z","signature_b64":"YFYs3A8S9b6nU+xwuBaWDvlmdyr1ASrgXNAmNBpmeXhE1OvixW0eDtCVnjlZtTzYK0QtpESa3/ZD7yG6jeFMAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bef3514cfe4482ae5b7dde8840e279031f9d0f7f7978f7fee43e36c8f5814f57","last_reissued_at":"2026-07-05T08:40:11.265934Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:40:11.265934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Read Between the Layers: Leveraging Multi-Layer Representations for Rehearsal-Free Continual Learning with Pre-Trained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Hans Hergen Lehmann, Jae Hee Lee, Kyra Ahrens, Stefan Wermter","submitted_at":"2023-12-13T13:11:44Z","abstract_excerpt":"We address the Continual Learning (CL) problem, wherein a model must learn a sequence of tasks from non-stationary distributions while preserving prior knowledge upon encountering new experiences. With the advancement of foundation models, CL research has pivoted from the initial learning-from-scratch paradigm towards utilizing generic features from large-scale pre-training. However, existing approaches to CL with pre-trained models primarily focus on separating class-specific features from the final representation layer and neglect the potential of intermediate representations to capture low-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.08888","kind":"arxiv","version":3},"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/2312.08888/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":"2312.08888","created_at":"2026-07-05T08:40:11.265996+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.08888v3","created_at":"2026-07-05T08:40:11.265996+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.08888","created_at":"2026-07-05T08:40:11.265996+00:00"},{"alias_kind":"pith_short_12","alias_value":"X3ZVCTH6ISBK","created_at":"2026-07-05T08:40:11.265996+00:00"},{"alias_kind":"pith_short_16","alias_value":"X3ZVCTH6ISBK4W35","created_at":"2026-07-05T08:40:11.265996+00:00"},{"alias_kind":"pith_short_8","alias_value":"X3ZVCTH6","created_at":"2026-07-05T08:40:11.265996+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.05387","citing_title":"The Generalization Ridge: Information Flow in Natural Language Generation","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM","json":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM.json","graph_json":"https://pith.science/api/pith-number/X3ZVCTH6ISBK4W3532EEBYTZAM/graph.json","events_json":"https://pith.science/api/pith-number/X3ZVCTH6ISBK4W3532EEBYTZAM/events.json","paper":"https://pith.science/paper/X3ZVCTH6"},"agent_actions":{"view_html":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM","download_json":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM.json","view_paper":"https://pith.science/paper/X3ZVCTH6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.08888&json=true","fetch_graph":"https://pith.science/api/pith-number/X3ZVCTH6ISBK4W3532EEBYTZAM/graph.json","fetch_events":"https://pith.science/api/pith-number/X3ZVCTH6ISBK4W3532EEBYTZAM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM/action/storage_attestation","attest_author":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM/action/author_attestation","sign_citation":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM/action/citation_signature","submit_replication":"https://pith.science/pith/X3ZVCTH6ISBK4W3532EEBYTZAM/action/replication_record"}},"created_at":"2026-07-05T08:40:11.265996+00:00","updated_at":"2026-07-05T08:40:11.265996+00:00"}