{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:Z2BXPRTYH4E7TW4B7PJQGJ7CFN","short_pith_number":"pith:Z2BXPRTY","schema_version":"1.0","canonical_sha256":"ce8377c6783f09f9db81fbd30327e22b69ca11705617f47188a2c8dd38b86258","source":{"kind":"arxiv","id":"2211.11031","version":5},"attestation_state":"computed","paper":{"title":"Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hamid Palangi, Marzyeh Ghassemi, Swami Sankaranarayanan, Thomas Hartvigsen, Yoon Kim","submitted_at":"2022-11-20T17:18:22Z","abstract_excerpt":"Deployed language models decay over time due to shifting inputs, changing user needs, or emergent world-knowledge gaps. When such problems are identified, we want to make targeted edits while avoiding expensive retraining. However, current model editors, which modify such behaviors of pre-trained models, degrade model performance quickly across multiple, sequential edits. We propose GRACE, a lifelong model editing method, which implements spot-fixes on streaming errors of a deployed model, ensuring minimal impact on unrelated inputs. GRACE writes new mappings into a pre-trained model's latent "},"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":"2211.11031","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-11-20T17:18:22Z","cross_cats_sorted":[],"title_canon_sha256":"3ca9edde34a042d7a63d36a36fdc229cd6079c6112bbc8ffc879672ae96a5c9a","abstract_canon_sha256":"db15a09f86a2993ab5ddda6e9e8e3ec663a84558facc645fc8ac87a7adec47cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:01:57.900396Z","signature_b64":"5CQ1PNmpqN15lNhz6qYCuxWxynRT8UJA+ziPkGxiQl0mEhuiJmYmlTv3ALvQnoaK8LuIaMqC1+FNtHlz0v3AAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce8377c6783f09f9db81fbd30327e22b69ca11705617f47188a2c8dd38b86258","last_reissued_at":"2026-07-05T07:01:57.899859Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:01:57.899859Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hamid Palangi, Marzyeh Ghassemi, Swami Sankaranarayanan, Thomas Hartvigsen, Yoon Kim","submitted_at":"2022-11-20T17:18:22Z","abstract_excerpt":"Deployed language models decay over time due to shifting inputs, changing user needs, or emergent world-knowledge gaps. When such problems are identified, we want to make targeted edits while avoiding expensive retraining. However, current model editors, which modify such behaviors of pre-trained models, degrade model performance quickly across multiple, sequential edits. We propose GRACE, a lifelong model editing method, which implements spot-fixes on streaming errors of a deployed model, ensuring minimal impact on unrelated inputs. GRACE writes new mappings into a pre-trained model's latent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.11031","kind":"arxiv","version":5},"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/2211.11031/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":"2211.11031","created_at":"2026-07-05T07:01:57.899915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.11031v5","created_at":"2026-07-05T07:01:57.899915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.11031","created_at":"2026-07-05T07:01:57.899915+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2BXPRTYH4E7","created_at":"2026-07-05T07:01:57.899915+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2BXPRTYH4E7TW4B","created_at":"2026-07-05T07:01:57.899915+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2BXPRTY","created_at":"2026-07-05T07:01:57.899915+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05310","citing_title":"Evaluating and Understanding Model Editing for Medical Vision Language Models","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26806","citing_title":"Memory Depth, Not Memory Access: Selective Parametric Consolidation for Long-Running Language Agents","ref_index":43,"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":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12841","citing_title":"TimeROME-DLM: Temporal Causal Tracing and Low-Rank Inference-Time Knowledge Editing for Masked Diffusion Language Models","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03924","citing_title":"Knowledge Editing in Masked Diffusion Language Models","ref_index":42,"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":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29916","citing_title":"EVAF: A Test-Retest Protocol for Selective Parametric Consolidation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23780","citing_title":"Beyond Binary Edits Robust Multimodal Knowledge Editing with Adversarial Subspace Alignment","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2510.09885","citing_title":"Diffusion-Inspired Masked Fine-Tuning for Knowledge Injection in Autoregressive LLMs","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02543","citing_title":"Norm Anchors Make Model Edits Last","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2507.03724","citing_title":"MemOS: A Memory OS for AI System","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19089","citing_title":"Towards Scalable Lifelong Knowledge Editing with Selective Knowledge Suppression","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN","json":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN.json","graph_json":"https://pith.science/api/pith-number/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/graph.json","events_json":"https://pith.science/api/pith-number/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/events.json","paper":"https://pith.science/paper/Z2BXPRTY"},"agent_actions":{"view_html":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN","download_json":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN.json","view_paper":"https://pith.science/paper/Z2BXPRTY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.11031&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/action/storage_attestation","attest_author":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/action/author_attestation","sign_citation":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/action/citation_signature","submit_replication":"https://pith.science/pith/Z2BXPRTYH4E7TW4B7PJQGJ7CFN/action/replication_record"}},"created_at":"2026-07-05T07:01:57.899915+00:00","updated_at":"2026-07-05T07:01:57.899915+00:00"}