{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KHDL22XHMSWSICAP2G6JRM2EJM","short_pith_number":"pith:KHDL22XH","schema_version":"1.0","canonical_sha256":"51c6bd6ae764ad24080fd1bc98b3444b323e8308a34e6c887bf31ee3ca712525","source":{"kind":"arxiv","id":"2405.17346","version":1},"attestation_state":"computed","paper":{"title":"Prompt Optimization with Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arun Verma, Bryan Kian Hsiang Low, Patrick Jaillet, See-kiong Ng, Xiaoqiang Lin, Zhongxiang Dai","submitted_at":"2024-05-27T16:49:29Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performances in various tasks. However, the performance of LLMs heavily depends on the input prompt, which has given rise to a number of recent works on prompt optimization. However, previous works often require the availability of a numeric score to assess the quality of every prompt. Unfortunately, when a human user interacts with a black-box LLM, attaining such a score is often infeasible and unreliable. Instead, it is usually significantly easier and more reliable to obtain preference feedback from a human user, i.e., showing the us"},"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":"2405.17346","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T16:49:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a0f2660531454ceacdca1e9d99d17c967ddf1fe8dc0e629d5dc5580e5d27f1b","abstract_canon_sha256":"8e96e209b58c6e0612c9a4e73169f34eb48d5c8314f76e54169b11d8fef882c8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:43.182260Z","signature_b64":"raFK1y3VkQ26jw6iItvKUjInRiwKTYwhHS4YbIEhfr23GpVdfSXReNLDG9mwkeNZWqt2RapL7ifn/i8UpTjsBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51c6bd6ae764ad24080fd1bc98b3444b323e8308a34e6c887bf31ee3ca712525","last_reissued_at":"2026-07-05T08:23:43.181775Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:43.181775Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt Optimization with Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arun Verma, Bryan Kian Hsiang Low, Patrick Jaillet, See-kiong Ng, Xiaoqiang Lin, Zhongxiang Dai","submitted_at":"2024-05-27T16:49:29Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performances in various tasks. However, the performance of LLMs heavily depends on the input prompt, which has given rise to a number of recent works on prompt optimization. However, previous works often require the availability of a numeric score to assess the quality of every prompt. Unfortunately, when a human user interacts with a black-box LLM, attaining such a score is often infeasible and unreliable. Instead, it is usually significantly easier and more reliable to obtain preference feedback from a human user, i.e., showing the us"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17346","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/2405.17346/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":"2405.17346","created_at":"2026-07-05T08:23:43.181833+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.17346v1","created_at":"2026-07-05T08:23:43.181833+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17346","created_at":"2026-07-05T08:23:43.181833+00:00"},{"alias_kind":"pith_short_12","alias_value":"KHDL22XHMSWS","created_at":"2026-07-05T08:23:43.181833+00:00"},{"alias_kind":"pith_short_16","alias_value":"KHDL22XHMSWSICAP","created_at":"2026-07-05T08:23:43.181833+00:00"},{"alias_kind":"pith_short_8","alias_value":"KHDL22XH","created_at":"2026-07-05T08:23:43.181833+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24004","citing_title":"Towards Spec Learning: Inference-Time Alignment from Preference Pairs","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24004","citing_title":"Towards Spec Learning: Inference-Time Alignment from Preference Pairs","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2508.07407","citing_title":"A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2512.13564","citing_title":"Memory in the Age of AI Agents","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19114","citing_title":"OOPrompt: Reifying Intents into Structured Artifacts for Modular and Iterative Prompting","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM","json":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM.json","graph_json":"https://pith.science/api/pith-number/KHDL22XHMSWSICAP2G6JRM2EJM/graph.json","events_json":"https://pith.science/api/pith-number/KHDL22XHMSWSICAP2G6JRM2EJM/events.json","paper":"https://pith.science/paper/KHDL22XH"},"agent_actions":{"view_html":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM","download_json":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM.json","view_paper":"https://pith.science/paper/KHDL22XH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.17346&json=true","fetch_graph":"https://pith.science/api/pith-number/KHDL22XHMSWSICAP2G6JRM2EJM/graph.json","fetch_events":"https://pith.science/api/pith-number/KHDL22XHMSWSICAP2G6JRM2EJM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM/action/storage_attestation","attest_author":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM/action/author_attestation","sign_citation":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM/action/citation_signature","submit_replication":"https://pith.science/pith/KHDL22XHMSWSICAP2G6JRM2EJM/action/replication_record"}},"created_at":"2026-07-05T08:23:43.181833+00:00","updated_at":"2026-07-05T08:23:43.181833+00:00"}