{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UAA7IU2RNOBCG3O56VOC7LAZ5Z","short_pith_number":"pith:UAA7IU2R","schema_version":"1.0","canonical_sha256":"a001f453516b82236dddf55c2fac19ee51d42f36d6a6dbf65fc353df6ea08c18","source":{"kind":"arxiv","id":"2505.09952","version":1},"attestation_state":"computed","paper":{"title":"Task-Core Memory Management and Consolidation for Long-term Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jie Zhou, Liang He, Qin Chen, Tianyu Huai, Wen Wu, Xingjiao Wu, Xipeng Qiu, Yuxuan Cai","submitted_at":"2025-05-15T04:22:35Z","abstract_excerpt":"In this paper, we focus on a long-term continual learning (CL) task, where a model learns sequentially from a stream of vast tasks over time, acquiring new knowledge while retaining previously learned information in a manner akin to human learning. Unlike traditional CL settings, long-term CL involves handling a significantly larger number of tasks, which exacerbates the issue of catastrophic forgetting. Our work seeks to address two critical questions: 1) How do existing CL methods perform in the context of long-term CL? and 2) How can we mitigate the catastrophic forgetting that arises from "},"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":"2505.09952","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-15T04:22:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5d7a421b8b8de8a9dfac9e3c8ab859c8a053ac6652d0ce92dc70e1269cc73968","abstract_canon_sha256":"71f9b1dbc298596941352bbbbe62db764f3546067773cb21f681495eb351d6de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:30.999459Z","signature_b64":"TsEH0NeimYmLuvIZ1iT3PTcfRM84c49TrbwrmKvWBY13AsBMEdcbAYF40X419o0URSW7M50hXuwWPXcl0KBQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a001f453516b82236dddf55c2fac19ee51d42f36d6a6dbf65fc353df6ea08c18","last_reissued_at":"2026-07-05T11:03:30.998986Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:30.998986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task-Core Memory Management and Consolidation for Long-term Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jie Zhou, Liang He, Qin Chen, Tianyu Huai, Wen Wu, Xingjiao Wu, Xipeng Qiu, Yuxuan Cai","submitted_at":"2025-05-15T04:22:35Z","abstract_excerpt":"In this paper, we focus on a long-term continual learning (CL) task, where a model learns sequentially from a stream of vast tasks over time, acquiring new knowledge while retaining previously learned information in a manner akin to human learning. Unlike traditional CL settings, long-term CL involves handling a significantly larger number of tasks, which exacerbates the issue of catastrophic forgetting. Our work seeks to address two critical questions: 1) How do existing CL methods perform in the context of long-term CL? and 2) How can we mitigate the catastrophic forgetting that arises from "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.09952","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/2505.09952/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":"2505.09952","created_at":"2026-07-05T11:03:30.999041+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.09952v1","created_at":"2026-07-05T11:03:30.999041+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.09952","created_at":"2026-07-05T11:03:30.999041+00:00"},{"alias_kind":"pith_short_12","alias_value":"UAA7IU2RNOBC","created_at":"2026-07-05T11:03:30.999041+00:00"},{"alias_kind":"pith_short_16","alias_value":"UAA7IU2RNOBCG3O5","created_at":"2026-07-05T11:03:30.999041+00:00"},{"alias_kind":"pith_short_8","alias_value":"UAA7IU2R","created_at":"2026-07-05T11:03:30.999041+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.16079","citing_title":"EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17183","citing_title":"LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2510.16079","citing_title":"EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00931","citing_title":"PsychAgent: An Experience-Driven Lifelong Learning Agent for Self-Evolving Psychological Counselor","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z","json":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z.json","graph_json":"https://pith.science/api/pith-number/UAA7IU2RNOBCG3O56VOC7LAZ5Z/graph.json","events_json":"https://pith.science/api/pith-number/UAA7IU2RNOBCG3O56VOC7LAZ5Z/events.json","paper":"https://pith.science/paper/UAA7IU2R"},"agent_actions":{"view_html":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z","download_json":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z.json","view_paper":"https://pith.science/paper/UAA7IU2R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.09952&json=true","fetch_graph":"https://pith.science/api/pith-number/UAA7IU2RNOBCG3O56VOC7LAZ5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/UAA7IU2RNOBCG3O56VOC7LAZ5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z/action/storage_attestation","attest_author":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z/action/author_attestation","sign_citation":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z/action/citation_signature","submit_replication":"https://pith.science/pith/UAA7IU2RNOBCG3O56VOC7LAZ5Z/action/replication_record"}},"created_at":"2026-07-05T11:03:30.999041+00:00","updated_at":"2026-07-05T11:03:30.999041+00:00"}