{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EUITVQY3YMY3BEIEYWHUMRGIC7","short_pith_number":"pith:EUITVQY3","schema_version":"1.0","canonical_sha256":"25113ac31bc331b09104c58f4644c817d4a576a8850efe955880bd659b9fa819","source":{"kind":"arxiv","id":"2506.13045","version":4},"attestation_state":"computed","paper":{"title":"Continual Learning for Generative AI: From LLMs to MLLMs and Beyond","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Cheng-Lin Liu, Da-Han Wang, Fanhu Zeng, Fei Zhu, Haiyang Guo, Hongbo Zhao, Jiayi Wang, Jingang Zhou, Shijie Ma, Wenzhuo Liu, Xukai Wang, Xu-Yao Zhang","submitted_at":"2025-06-16T02:27:25Z","abstract_excerpt":"The rapid advancement of generative models has empowered modern AI systems to comprehend and produce highly sophisticated content, even achieving human-level performance in specific domains. However, these models are fundamentally constrained by \\emph{catastrophic forgetting}, \\ie~a persistent challenge where models experience performance degradation on previously learned tasks when adapting to new tasks. To address this practical limitation, numerous approaches have been proposed to enhance the adaptability and scalability of generative AI in real-world applications. In this work, we present "},"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":"2506.13045","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T02:27:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"eaaa352a984f225a9071e18252d35d2a87aafad3f883b916b7aa98a75c732e92","abstract_canon_sha256":"150c0229e45b79ccd336ac4d9e3eada5178c1d846d4d036802daba80a7ce18bd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:58:31.414373Z","signature_b64":"9DVqVHHhP/tHusB5rjAnWcovCOY3iB2RkDzGTLqtNhr5t4xJcsl7Uua9dGRCppnb14LSeICHjrntd/DfZJOSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25113ac31bc331b09104c58f4644c817d4a576a8850efe955880bd659b9fa819","last_reissued_at":"2026-07-05T11:58:31.413908Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:58:31.413908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continual Learning for Generative AI: From LLMs to MLLMs and Beyond","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Cheng-Lin Liu, Da-Han Wang, Fanhu Zeng, Fei Zhu, Haiyang Guo, Hongbo Zhao, Jiayi Wang, Jingang Zhou, Shijie Ma, Wenzhuo Liu, Xukai Wang, Xu-Yao Zhang","submitted_at":"2025-06-16T02:27:25Z","abstract_excerpt":"The rapid advancement of generative models has empowered modern AI systems to comprehend and produce highly sophisticated content, even achieving human-level performance in specific domains. However, these models are fundamentally constrained by \\emph{catastrophic forgetting}, \\ie~a persistent challenge where models experience performance degradation on previously learned tasks when adapting to new tasks. To address this practical limitation, numerous approaches have been proposed to enhance the adaptability and scalability of generative AI in real-world applications. In this work, we present "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13045","kind":"arxiv","version":4},"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/2506.13045/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":"2506.13045","created_at":"2026-07-05T11:58:31.413965+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.13045v4","created_at":"2026-07-05T11:58:31.413965+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13045","created_at":"2026-07-05T11:58:31.413965+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUITVQY3YMY3","created_at":"2026-07-05T11:58:31.413965+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUITVQY3YMY3BEIE","created_at":"2026-07-05T11:58:31.413965+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUITVQY3","created_at":"2026-07-05T11:58:31.413965+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30266","citing_title":"Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2510.20685","citing_title":"C-NAV: Towards Self-Evolving Continual Object Navigation in Open World","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10765","citing_title":"Dynamic Cross-Modal Prompt Generation for Multimodal Continual Instruction Tuning","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7","json":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7.json","graph_json":"https://pith.science/api/pith-number/EUITVQY3YMY3BEIEYWHUMRGIC7/graph.json","events_json":"https://pith.science/api/pith-number/EUITVQY3YMY3BEIEYWHUMRGIC7/events.json","paper":"https://pith.science/paper/EUITVQY3"},"agent_actions":{"view_html":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7","download_json":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7.json","view_paper":"https://pith.science/paper/EUITVQY3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.13045&json=true","fetch_graph":"https://pith.science/api/pith-number/EUITVQY3YMY3BEIEYWHUMRGIC7/graph.json","fetch_events":"https://pith.science/api/pith-number/EUITVQY3YMY3BEIEYWHUMRGIC7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7/action/storage_attestation","attest_author":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7/action/author_attestation","sign_citation":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7/action/citation_signature","submit_replication":"https://pith.science/pith/EUITVQY3YMY3BEIEYWHUMRGIC7/action/replication_record"}},"created_at":"2026-07-05T11:58:31.413965+00:00","updated_at":"2026-07-05T11:58:31.413965+00:00"}