{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YMTD3GGPDAMX2ZVENVNXRSBGJZ","short_pith_number":"pith:YMTD3GGP","schema_version":"1.0","canonical_sha256":"c3263d98cf18197d66a46d5b78c8264e4a4f917b618ee81a624a7b644e4be9fc","source":{"kind":"arxiv","id":"2310.20348","version":1},"attestation_state":"computed","paper":{"title":"Class Incremental Learning with Pre-trained Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrew D. Bagdanov, Haori Lu, Jia-Wen Xiao, Ming-Ming Cheng, Xialei Liu, Xusheng Cao","submitted_at":"2023-10-31T10:45:03Z","abstract_excerpt":"With the advent of large-scale pre-trained models, interest in adapting and exploiting them for continual learning scenarios has grown.\n  In this paper, we propose an approach to exploiting pre-trained vision-language models (e.g. CLIP) that enables further adaptation instead of only using zero-shot learning of new tasks. We augment a pre-trained CLIP model with additional layers after the Image Encoder or before the Text Encoder. We investigate three different strategies: a Linear Adapter, a Self-attention Adapter, each operating on the image embedding, and Prompt Tuning which instead modifie"},"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":"2310.20348","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-10-31T10:45:03Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1a77b14946c98c9a51bab92fe921624035b9854248cbb87d35c05fe20917c8d5","abstract_canon_sha256":"d9679144d90ce262fddf8633d647442465fe37bf9d94294ae18bdbc29a92c517"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:07:28.881737Z","signature_b64":"JYdJRwkfrvNYsdK6W0EtK74EtRQ/0HjqmDOXpjKdvWL5zdL70kMOMynssIBskGPt68xVW8y3sni3QpBaOMKnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3263d98cf18197d66a46d5b78c8264e4a4f917b618ee81a624a7b644e4be9fc","last_reissued_at":"2026-07-05T07:07:28.881250Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:07:28.881250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Class Incremental Learning with Pre-trained Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrew D. Bagdanov, Haori Lu, Jia-Wen Xiao, Ming-Ming Cheng, Xialei Liu, Xusheng Cao","submitted_at":"2023-10-31T10:45:03Z","abstract_excerpt":"With the advent of large-scale pre-trained models, interest in adapting and exploiting them for continual learning scenarios has grown.\n  In this paper, we propose an approach to exploiting pre-trained vision-language models (e.g. CLIP) that enables further adaptation instead of only using zero-shot learning of new tasks. We augment a pre-trained CLIP model with additional layers after the Image Encoder or before the Text Encoder. We investigate three different strategies: a Linear Adapter, a Self-attention Adapter, each operating on the image embedding, and Prompt Tuning which instead modifie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.20348","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/2310.20348/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":"2310.20348","created_at":"2026-07-05T07:07:28.881307+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.20348v1","created_at":"2026-07-05T07:07:28.881307+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.20348","created_at":"2026-07-05T07:07:28.881307+00:00"},{"alias_kind":"pith_short_12","alias_value":"YMTD3GGPDAMX","created_at":"2026-07-05T07:07:28.881307+00:00"},{"alias_kind":"pith_short_16","alias_value":"YMTD3GGPDAMX2ZVE","created_at":"2026-07-05T07:07:28.881307+00:00"},{"alias_kind":"pith_short_8","alias_value":"YMTD3GGP","created_at":"2026-07-05T07:07:28.881307+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.04227","citing_title":"Continual Learning for VLMs: A Survey and Taxonomy Beyond Forgetting","ref_index":101,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19301","citing_title":"iGSP:Implicit Gradient Subspace Projection for Efficient Continual Learning of Vision-Language Models","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ","json":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ.json","graph_json":"https://pith.science/api/pith-number/YMTD3GGPDAMX2ZVENVNXRSBGJZ/graph.json","events_json":"https://pith.science/api/pith-number/YMTD3GGPDAMX2ZVENVNXRSBGJZ/events.json","paper":"https://pith.science/paper/YMTD3GGP"},"agent_actions":{"view_html":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ","download_json":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ.json","view_paper":"https://pith.science/paper/YMTD3GGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.20348&json=true","fetch_graph":"https://pith.science/api/pith-number/YMTD3GGPDAMX2ZVENVNXRSBGJZ/graph.json","fetch_events":"https://pith.science/api/pith-number/YMTD3GGPDAMX2ZVENVNXRSBGJZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ/action/storage_attestation","attest_author":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ/action/author_attestation","sign_citation":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ/action/citation_signature","submit_replication":"https://pith.science/pith/YMTD3GGPDAMX2ZVENVNXRSBGJZ/action/replication_record"}},"created_at":"2026-07-05T07:07:28.881307+00:00","updated_at":"2026-07-05T07:07:28.881307+00:00"}