{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GAVNL6QR3S4TPFI3EB6BS6FNQN","short_pith_number":"pith:GAVNL6QR","schema_version":"1.0","canonical_sha256":"302ad5fa11dcb937951b207c1978ad8348c7d6b6f0b7e8e6ab54f5b30734c67e","source":{"kind":"arxiv","id":"2506.09672","version":1},"attestation_state":"computed","paper":{"title":"Is Fine-Tuning an Effective Solution? Reassessing Knowledge Editing for Unstructured Data","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chuanyuan Tan, Hao Xiong, Wenliang Chen","submitted_at":"2025-06-11T12:43:10Z","abstract_excerpt":"Unstructured Knowledge Editing (UKE) is crucial for updating the relevant knowledge of large language models (LLMs). It focuses on unstructured inputs, such as long or free-form texts, which are common forms of real-world knowledge. Although previous studies have proposed effective methods and tested them, some issues exist: (1) Lack of Locality evaluation for UKE, and (2) Abnormal failure of fine-tuning (FT) based methods for UKE. To address these issues, we first construct two datasets, UnKEBench-Loc and AKEW-Loc (CF), by extending two existing UKE datasets with locality test data from the u"},"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":"2506.09672","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T12:43:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fba273bc25cc812aea92570f0e1a2bdd37cab2c456c8c08d8b4aa0aec4078512","abstract_canon_sha256":"71f4a33569e10e50955da99dca7e0688f0cc8015644253a48bef27376223fbb5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:55.122183Z","signature_b64":"kr5Cq/m3xL4PMX8XYansjJPlGSl01wukAO5WnTXNe8G7XVWPuza39zMDMhJeM4jjO+UiqP2Bcm4GA8rgLpCDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"302ad5fa11dcb937951b207c1978ad8348c7d6b6f0b7e8e6ab54f5b30734c67e","last_reissued_at":"2026-07-05T11:19:55.121669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:55.121669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Fine-Tuning an Effective Solution? Reassessing Knowledge Editing for Unstructured Data","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chuanyuan Tan, Hao Xiong, Wenliang Chen","submitted_at":"2025-06-11T12:43:10Z","abstract_excerpt":"Unstructured Knowledge Editing (UKE) is crucial for updating the relevant knowledge of large language models (LLMs). It focuses on unstructured inputs, such as long or free-form texts, which are common forms of real-world knowledge. Although previous studies have proposed effective methods and tested them, some issues exist: (1) Lack of Locality evaluation for UKE, and (2) Abnormal failure of fine-tuning (FT) based methods for UKE. To address these issues, we first construct two datasets, UnKEBench-Loc and AKEW-Loc (CF), by extending two existing UKE datasets with locality test data from the u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09672","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/2506.09672/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":"2506.09672","created_at":"2026-07-05T11:19:55.121738+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.09672v1","created_at":"2026-07-05T11:19:55.121738+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09672","created_at":"2026-07-05T11:19:55.121738+00:00"},{"alias_kind":"pith_short_12","alias_value":"GAVNL6QR3S4T","created_at":"2026-07-05T11:19:55.121738+00:00"},{"alias_kind":"pith_short_16","alias_value":"GAVNL6QR3S4TPFI3","created_at":"2026-07-05T11:19:55.121738+00:00"},{"alias_kind":"pith_short_8","alias_value":"GAVNL6QR","created_at":"2026-07-05T11:19:55.121738+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.14242","citing_title":"Implicit Reasoning Steering via Concept Chaining","ref_index":177,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN","json":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN.json","graph_json":"https://pith.science/api/pith-number/GAVNL6QR3S4TPFI3EB6BS6FNQN/graph.json","events_json":"https://pith.science/api/pith-number/GAVNL6QR3S4TPFI3EB6BS6FNQN/events.json","paper":"https://pith.science/paper/GAVNL6QR"},"agent_actions":{"view_html":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN","download_json":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN.json","view_paper":"https://pith.science/paper/GAVNL6QR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.09672&json=true","fetch_graph":"https://pith.science/api/pith-number/GAVNL6QR3S4TPFI3EB6BS6FNQN/graph.json","fetch_events":"https://pith.science/api/pith-number/GAVNL6QR3S4TPFI3EB6BS6FNQN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN/action/storage_attestation","attest_author":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN/action/author_attestation","sign_citation":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN/action/citation_signature","submit_replication":"https://pith.science/pith/GAVNL6QR3S4TPFI3EB6BS6FNQN/action/replication_record"}},"created_at":"2026-07-05T11:19:55.121738+00:00","updated_at":"2026-07-05T11:19:55.121738+00:00"}