{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6QCR3CJDXDUQOW6ANCAMYROJ7F","short_pith_number":"pith:6QCR3CJD","schema_version":"1.0","canonical_sha256":"f4051d8923b8e9075bc06880cc45c9f940c29e69ecf5cc27c4517c3c7085ddca","source":{"kind":"arxiv","id":"2311.09510","version":3},"attestation_state":"computed","paper":{"title":"Tailoring with Targeted Precision: Edit-Based Agents for Open-Domain Procedure Customization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bodhisattwa Prasad Majumder, Faeze Brahman, Li Zhang, Niket Tandon, Peter Clark, Yash Kumar Lal","submitted_at":"2023-11-16T02:25:36Z","abstract_excerpt":"How-to procedures, such as how to plant a garden, are now used by millions of users, but sometimes need customizing to meet a user's specific needs, e.g., planting a garden without pesticides. Our goal is to measure and improve an LLM's ability to perform such customization. Our approach is to test several simple multi-LLM-agent architectures for customization, as well as an end-to-end LLM, using a new evaluation set, called CustomPlans, of over 200 WikiHow procedures each with a customization need. We find that a simple architecture with two LLM agents used sequentially performs best, one tha"},"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":"2311.09510","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-16T02:25:36Z","cross_cats_sorted":[],"title_canon_sha256":"544fd8bcfe443af60ddad277ba3a611c7b6984353261080c7bd58790a6f7d33d","abstract_canon_sha256":"90b9bc3491629722f4154b9aa9b2b1a74e07e4280c69d312a2650b7196e7e5cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:29.834125Z","signature_b64":"xpATIWzPALmR5bs9Dgeq/h0P7a1e9HzH7Z/nPs/KgZmLh8z6a7MbhMytz+y1yGu6d+3+2hFRlLDZuvYrKnp1AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4051d8923b8e9075bc06880cc45c9f940c29e69ecf5cc27c4517c3c7085ddca","last_reissued_at":"2026-07-05T08:25:29.833611Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:29.833611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tailoring with Targeted Precision: Edit-Based Agents for Open-Domain Procedure Customization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bodhisattwa Prasad Majumder, Faeze Brahman, Li Zhang, Niket Tandon, Peter Clark, Yash Kumar Lal","submitted_at":"2023-11-16T02:25:36Z","abstract_excerpt":"How-to procedures, such as how to plant a garden, are now used by millions of users, but sometimes need customizing to meet a user's specific needs, e.g., planting a garden without pesticides. Our goal is to measure and improve an LLM's ability to perform such customization. Our approach is to test several simple multi-LLM-agent architectures for customization, as well as an end-to-end LLM, using a new evaluation set, called CustomPlans, of over 200 WikiHow procedures each with a customization need. We find that a simple architecture with two LLM agents used sequentially performs best, one tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.09510","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/2311.09510/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":"2311.09510","created_at":"2026-07-05T08:25:29.833677+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.09510v3","created_at":"2026-07-05T08:25:29.833677+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.09510","created_at":"2026-07-05T08:25:29.833677+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QCR3CJDXDUQ","created_at":"2026-07-05T08:25:29.833677+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QCR3CJDXDUQOW6A","created_at":"2026-07-05T08:25:29.833677+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QCR3CJD","created_at":"2026-07-05T08:25:29.833677+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.12626","citing_title":"DoubleAgents: Human-Agent Alignment in a Socially Embedded Workflow","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F","json":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F.json","graph_json":"https://pith.science/api/pith-number/6QCR3CJDXDUQOW6ANCAMYROJ7F/graph.json","events_json":"https://pith.science/api/pith-number/6QCR3CJDXDUQOW6ANCAMYROJ7F/events.json","paper":"https://pith.science/paper/6QCR3CJD"},"agent_actions":{"view_html":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F","download_json":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F.json","view_paper":"https://pith.science/paper/6QCR3CJD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.09510&json=true","fetch_graph":"https://pith.science/api/pith-number/6QCR3CJDXDUQOW6ANCAMYROJ7F/graph.json","fetch_events":"https://pith.science/api/pith-number/6QCR3CJDXDUQOW6ANCAMYROJ7F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F/action/storage_attestation","attest_author":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F/action/author_attestation","sign_citation":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F/action/citation_signature","submit_replication":"https://pith.science/pith/6QCR3CJDXDUQOW6ANCAMYROJ7F/action/replication_record"}},"created_at":"2026-07-05T08:25:29.833677+00:00","updated_at":"2026-07-05T08:25:29.833677+00:00"}