{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BIZZE6X62F5KPCB4GC5O45VPMN","short_pith_number":"pith:BIZZE6X6","schema_version":"1.0","canonical_sha256":"0a33927afed17aa7883c30baee76af634df626679773c811f734befde0e02322","source":{"kind":"arxiv","id":"2404.15269","version":3},"attestation_state":"computed","paper":{"title":"Aligning LLM Agents by Learning Latent Preference from User Edits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexey Taymanov, Dipendra Misra, Eduardo Salinas, Ge Gao, Paul Mineiro","submitted_at":"2024-04-23T17:57:47Z","abstract_excerpt":"We study interactive learning of LLM-based language agents based on user edits made to the agent's output. In a typical setting such as writing assistants, the user interacts with a language agent to generate a response given a context, and may optionally edit the agent response to personalize it based on their latent preference, in addition to improving the correctness. The edit feedback is naturally generated, making it a suitable candidate for improving the agent's alignment with the user's preference, and for reducing the cost of user edits over time. We propose a learning framework, PRELU"},"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":"2404.15269","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-23T17:57:47Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"d84fe8257488474a432ba7ea656b6c8e341cb22802f398712423a4b11485875f","abstract_canon_sha256":"6905b11b5057fec3417cd2430af132bad56b10c4168e432a5b03fa6b522089d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:39:22.706472Z","signature_b64":"W0iyNPIAMx1YSrQyQZSEAxk+TLL3Ep1VT2OzmRLl2lGreieg1e6S8FXLZDCLwilyuEc8TZBVlYh/1VtPhHZxCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a33927afed17aa7883c30baee76af634df626679773c811f734befde0e02322","last_reissued_at":"2026-07-05T09:39:22.705947Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:39:22.705947Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aligning LLM Agents by Learning Latent Preference from User Edits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexey Taymanov, Dipendra Misra, Eduardo Salinas, Ge Gao, Paul Mineiro","submitted_at":"2024-04-23T17:57:47Z","abstract_excerpt":"We study interactive learning of LLM-based language agents based on user edits made to the agent's output. In a typical setting such as writing assistants, the user interacts with a language agent to generate a response given a context, and may optionally edit the agent response to personalize it based on their latent preference, in addition to improving the correctness. The edit feedback is naturally generated, making it a suitable candidate for improving the agent's alignment with the user's preference, and for reducing the cost of user edits over time. We propose a learning framework, PRELU"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.15269","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/2404.15269/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":"2404.15269","created_at":"2026-07-05T09:39:22.706013+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.15269v3","created_at":"2026-07-05T09:39:22.706013+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.15269","created_at":"2026-07-05T09:39:22.706013+00:00"},{"alias_kind":"pith_short_12","alias_value":"BIZZE6X62F5K","created_at":"2026-07-05T09:39:22.706013+00:00"},{"alias_kind":"pith_short_16","alias_value":"BIZZE6X62F5KPCB4","created_at":"2026-07-05T09:39:22.706013+00:00"},{"alias_kind":"pith_short_8","alias_value":"BIZZE6X6","created_at":"2026-07-05T09:39:22.706013+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.06352","citing_title":"CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04901","citing_title":"FileGram: Grounding Agent Personalization in File-System Behavioral Traces","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN","json":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN.json","graph_json":"https://pith.science/api/pith-number/BIZZE6X62F5KPCB4GC5O45VPMN/graph.json","events_json":"https://pith.science/api/pith-number/BIZZE6X62F5KPCB4GC5O45VPMN/events.json","paper":"https://pith.science/paper/BIZZE6X6"},"agent_actions":{"view_html":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN","download_json":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN.json","view_paper":"https://pith.science/paper/BIZZE6X6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.15269&json=true","fetch_graph":"https://pith.science/api/pith-number/BIZZE6X62F5KPCB4GC5O45VPMN/graph.json","fetch_events":"https://pith.science/api/pith-number/BIZZE6X62F5KPCB4GC5O45VPMN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN/action/storage_attestation","attest_author":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN/action/author_attestation","sign_citation":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN/action/citation_signature","submit_replication":"https://pith.science/pith/BIZZE6X62F5KPCB4GC5O45VPMN/action/replication_record"}},"created_at":"2026-07-05T09:39:22.706013+00:00","updated_at":"2026-07-05T09:39:22.706013+00:00"}