{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GWEUPL5XHIN2BI7PXNNGNYAXQ4","short_pith_number":"pith:GWEUPL5X","schema_version":"1.0","canonical_sha256":"358947afb73a1ba0a3efbb5a66e017872efdd64417110e60c3f90676668ae64b","source":{"kind":"arxiv","id":"2504.12408","version":1},"attestation_state":"computed","paper":{"title":"A Human-AI Comparative Analysis of Prompt Sensitivity in LLM-Based Relevance Judgment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Charles L. A . Clarke, Negar Arabzadeh","submitted_at":"2025-04-16T18:17:19Z","abstract_excerpt":"Large Language Models (LLMs) are increasingly used to automate relevance judgments for information retrieval (IR) tasks, often demonstrating agreement with human labels that approaches inter-human agreement. To assess the robustness and reliability of LLM-based relevance judgments, we systematically investigate impact of prompt sensitivity on the task. We collected prompts for relevance assessment from 15 human experts and 15 LLMs across three tasks~ -- ~binary, graded, and pairwise~ -- ~yielding 90 prompts in total. After filtering out unusable prompts from three humans and three LLMs, we emp"},"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":"2504.12408","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-04-16T18:17:19Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"6dd9af01595d793c478d75b69240c1f02278560e299b54f8482357d19cd90a2e","abstract_canon_sha256":"64c87e1d9fd05edfad2dce99ffbd4484134f0c80ef1998735089a61834409946"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:15.917908Z","signature_b64":"Yb5rKnTHS8R6ChJdH66nq2zOOByFjJkcA/crWLyWYCPV6CGGvUiHPT+XgUq8XABIG759+D4QhoABLrgprtSjBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"358947afb73a1ba0a3efbb5a66e017872efdd64417110e60c3f90676668ae64b","last_reissued_at":"2026-07-05T10:50:15.917419Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:15.917419Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Human-AI Comparative Analysis of Prompt Sensitivity in LLM-Based Relevance Judgment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Charles L. A . Clarke, Negar Arabzadeh","submitted_at":"2025-04-16T18:17:19Z","abstract_excerpt":"Large Language Models (LLMs) are increasingly used to automate relevance judgments for information retrieval (IR) tasks, often demonstrating agreement with human labels that approaches inter-human agreement. To assess the robustness and reliability of LLM-based relevance judgments, we systematically investigate impact of prompt sensitivity on the task. We collected prompts for relevance assessment from 15 human experts and 15 LLMs across three tasks~ -- ~binary, graded, and pairwise~ -- ~yielding 90 prompts in total. After filtering out unusable prompts from three humans and three LLMs, we emp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12408","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/2504.12408/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":"2504.12408","created_at":"2026-07-05T10:50:15.917478+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12408v1","created_at":"2026-07-05T10:50:15.917478+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12408","created_at":"2026-07-05T10:50:15.917478+00:00"},{"alias_kind":"pith_short_12","alias_value":"GWEUPL5XHIN2","created_at":"2026-07-05T10:50:15.917478+00:00"},{"alias_kind":"pith_short_16","alias_value":"GWEUPL5XHIN2BI7P","created_at":"2026-07-05T10:50:15.917478+00:00"},{"alias_kind":"pith_short_8","alias_value":"GWEUPL5X","created_at":"2026-07-05T10:50:15.917478+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.10477","citing_title":"PEEM: Prompt Engineering Evaluation Metrics for Interpretable Joint Evaluation of Prompts and Responses","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4","json":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4.json","graph_json":"https://pith.science/api/pith-number/GWEUPL5XHIN2BI7PXNNGNYAXQ4/graph.json","events_json":"https://pith.science/api/pith-number/GWEUPL5XHIN2BI7PXNNGNYAXQ4/events.json","paper":"https://pith.science/paper/GWEUPL5X"},"agent_actions":{"view_html":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4","download_json":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4.json","view_paper":"https://pith.science/paper/GWEUPL5X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12408&json=true","fetch_graph":"https://pith.science/api/pith-number/GWEUPL5XHIN2BI7PXNNGNYAXQ4/graph.json","fetch_events":"https://pith.science/api/pith-number/GWEUPL5XHIN2BI7PXNNGNYAXQ4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4/action/storage_attestation","attest_author":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4/action/author_attestation","sign_citation":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4/action/citation_signature","submit_replication":"https://pith.science/pith/GWEUPL5XHIN2BI7PXNNGNYAXQ4/action/replication_record"}},"created_at":"2026-07-05T10:50:15.917478+00:00","updated_at":"2026-07-05T10:50:15.917478+00:00"}