{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZE5BBZXWHJZA7FH453BJ3K7PB7","short_pith_number":"pith:ZE5BBZXW","schema_version":"1.0","canonical_sha256":"c93a10e6f63a720f94fceec29dabef0fcddbd7deaf87a2f738ffb2006831c765","source":{"kind":"arxiv","id":"2311.08011","version":2},"attestation_state":"computed","paper":{"title":"Forgetting before Learning: Utilizing Parametric Arithmetic for Knowledge Updating in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengming Li, Dingwei Chen, Min Yang, Ruifeng Xu, Shiwen Ni, Xiping Hu","submitted_at":"2023-11-14T09:12:40Z","abstract_excerpt":"Recent advancements in Large Language Models (LLMs) have showcased their remarkable capabilities in text understanding and generation. However, even stronger LLMs are susceptible to acquiring erroneous or obsolete information from the training corpus. Direct secondary fine-tuning with data containing new knowledge may be ineffective in updating knowledge due to the conflict between old and new knowledge. In this paper, we propose a new paradigm for fine-tuning called F-Learning (Forgetting before Learning), which employs parametric arithmetic to facilitate the forgetting of old knowledge and l"},"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":"2311.08011","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T09:12:40Z","cross_cats_sorted":[],"title_canon_sha256":"619a17e3b5c1e64898cd92db88ea6006f0ab130c69b22209c6dc8d82413f6d68","abstract_canon_sha256":"fccf001b3d4597ea5412614bbbb73e7e9870855b2f94a8d5d25d2007295af2ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:45:50.300716Z","signature_b64":"yfst6A8mJAraqvdwtRbbceaLRpu8Og1hbjnF08w1tJx3FGNcA3e1TQCWlpuyLRydeWZMx0zFw3fRcW836yRABQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c93a10e6f63a720f94fceec29dabef0fcddbd7deaf87a2f738ffb2006831c765","last_reissued_at":"2026-07-05T07:45:50.300166Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:45:50.300166Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Forgetting before Learning: Utilizing Parametric Arithmetic for Knowledge Updating in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengming Li, Dingwei Chen, Min Yang, Ruifeng Xu, Shiwen Ni, Xiping Hu","submitted_at":"2023-11-14T09:12:40Z","abstract_excerpt":"Recent advancements in Large Language Models (LLMs) have showcased their remarkable capabilities in text understanding and generation. However, even stronger LLMs are susceptible to acquiring erroneous or obsolete information from the training corpus. Direct secondary fine-tuning with data containing new knowledge may be ineffective in updating knowledge due to the conflict between old and new knowledge. In this paper, we propose a new paradigm for fine-tuning called F-Learning (Forgetting before Learning), which employs parametric arithmetic to facilitate the forgetting of old knowledge and l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08011","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/2311.08011/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":"2311.08011","created_at":"2026-07-05T07:45:50.300236+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.08011v2","created_at":"2026-07-05T07:45:50.300236+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08011","created_at":"2026-07-05T07:45:50.300236+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZE5BBZXWHJZA","created_at":"2026-07-05T07:45:50.300236+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZE5BBZXWHJZA7FH4","created_at":"2026-07-05T07:45:50.300236+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZE5BBZXW","created_at":"2026-07-05T07:45:50.300236+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.07666","citing_title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","ref_index":159,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7","json":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7.json","graph_json":"https://pith.science/api/pith-number/ZE5BBZXWHJZA7FH453BJ3K7PB7/graph.json","events_json":"https://pith.science/api/pith-number/ZE5BBZXWHJZA7FH453BJ3K7PB7/events.json","paper":"https://pith.science/paper/ZE5BBZXW"},"agent_actions":{"view_html":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7","download_json":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7.json","view_paper":"https://pith.science/paper/ZE5BBZXW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.08011&json=true","fetch_graph":"https://pith.science/api/pith-number/ZE5BBZXWHJZA7FH453BJ3K7PB7/graph.json","fetch_events":"https://pith.science/api/pith-number/ZE5BBZXWHJZA7FH453BJ3K7PB7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7/action/storage_attestation","attest_author":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7/action/author_attestation","sign_citation":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7/action/citation_signature","submit_replication":"https://pith.science/pith/ZE5BBZXWHJZA7FH453BJ3K7PB7/action/replication_record"}},"created_at":"2026-07-05T07:45:50.300236+00:00","updated_at":"2026-07-05T07:45:50.300236+00:00"}