{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PI4J4F2LLQ22OY52CA5WV2UQVR","short_pith_number":"pith:PI4J4F2L","schema_version":"1.0","canonical_sha256":"7a389e174b5c35a763ba103b6aea90ac54f0c7a8ec8e7c9250c7ebf627a90f90","source":{"kind":"arxiv","id":"2311.04661","version":3},"attestation_state":"computed","paper":{"title":"Massive Editing for Large Language Models via Meta Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chenmien Tan, Ge Zhang, Jie Fu","submitted_at":"2023-11-08T13:03:06Z","abstract_excerpt":"While large language models (LLMs) have enabled learning knowledge from the pre-training corpora, the acquired knowledge may be fundamentally incorrect or outdated over time, which necessitates rectifying the knowledge of the language model (LM) after the training. A promising approach involves employing a hyper-network to generate parameter shift, whereas existing hyper-networks suffer from inferior scalability in synchronous editing operation amount. To mitigate the problem, we propose the MAssive Language Model Editing Network (MALMEN), which formulates the parameter shift aggregation as th"},"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":"2311.04661","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-08T13:03:06Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cda5775d04cc93ad0a0bddfac83019051905dcb55c441e69380df27541b2e6ea","abstract_canon_sha256":"cee879428a692693ebb31ad186abfa5fb66e9bdd4814ef9c01632e7664f912ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:37:14.126240Z","signature_b64":"gks1JOgnt7mFCyS92SJGUifWRe6bszPQbo6RVvpOITzKi2+IsmFi3eqM1snfFbyuvys+8OA0dX7ZKy8K/4ITAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a389e174b5c35a763ba103b6aea90ac54f0c7a8ec8e7c9250c7ebf627a90f90","last_reissued_at":"2026-07-05T07:37:14.125851Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:37:14.125851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Massive Editing for Large Language Models via Meta Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chenmien Tan, Ge Zhang, Jie Fu","submitted_at":"2023-11-08T13:03:06Z","abstract_excerpt":"While large language models (LLMs) have enabled learning knowledge from the pre-training corpora, the acquired knowledge may be fundamentally incorrect or outdated over time, which necessitates rectifying the knowledge of the language model (LM) after the training. A promising approach involves employing a hyper-network to generate parameter shift, whereas existing hyper-networks suffer from inferior scalability in synchronous editing operation amount. To mitigate the problem, we propose the MAssive Language Model Editing Network (MALMEN), which formulates the parameter shift aggregation as th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.04661","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/2311.04661/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":"2311.04661","created_at":"2026-07-05T07:37:14.125908+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.04661v3","created_at":"2026-07-05T07:37:14.125908+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.04661","created_at":"2026-07-05T07:37:14.125908+00:00"},{"alias_kind":"pith_short_12","alias_value":"PI4J4F2LLQ22","created_at":"2026-07-05T07:37:14.125908+00:00"},{"alias_kind":"pith_short_16","alias_value":"PI4J4F2LLQ22OY52","created_at":"2026-07-05T07:37:14.125908+00:00"},{"alias_kind":"pith_short_8","alias_value":"PI4J4F2L","created_at":"2026-07-05T07:37:14.125908+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.14668","citing_title":"When to Write and When to Suppress: Route-Specialized Dual Adapters for Memory-Assisted Knowledge Editing","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.14668","citing_title":"When to Write and When to Suppress: Route-Specialized Dual Adapters for Memory-Assisted Knowledge Editing","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23780","citing_title":"Beyond Binary Edits Robust Multimodal Knowledge Editing with Adversarial Subspace Alignment","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19089","citing_title":"Towards Scalable Lifelong Knowledge Editing with Selective Knowledge Suppression","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR","json":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR.json","graph_json":"https://pith.science/api/pith-number/PI4J4F2LLQ22OY52CA5WV2UQVR/graph.json","events_json":"https://pith.science/api/pith-number/PI4J4F2LLQ22OY52CA5WV2UQVR/events.json","paper":"https://pith.science/paper/PI4J4F2L"},"agent_actions":{"view_html":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR","download_json":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR.json","view_paper":"https://pith.science/paper/PI4J4F2L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.04661&json=true","fetch_graph":"https://pith.science/api/pith-number/PI4J4F2LLQ22OY52CA5WV2UQVR/graph.json","fetch_events":"https://pith.science/api/pith-number/PI4J4F2LLQ22OY52CA5WV2UQVR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR/action/storage_attestation","attest_author":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR/action/author_attestation","sign_citation":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR/action/citation_signature","submit_replication":"https://pith.science/pith/PI4J4F2LLQ22OY52CA5WV2UQVR/action/replication_record"}},"created_at":"2026-07-05T07:37:14.125908+00:00","updated_at":"2026-07-05T07:37:14.125908+00:00"}