{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:U6UKN3K4PHJ2WSLR25TPVQOER2","short_pith_number":"pith:U6UKN3K4","schema_version":"1.0","canonical_sha256":"a7a8a6ed5c79d3ab4971d766fac1c48e9fbc77df692e9477fe1a902dc444c773","source":{"kind":"arxiv","id":"2402.15754","version":1},"attestation_state":"computed","paper":{"title":"HD-Eval: Aligning Large Language Model Evaluators Through Hierarchical Criteria Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, Qi Zhang, Shaohan Huang, Tianchi Yang, Weiwei Deng, Yuxuan Liu, Zihan Zhang","submitted_at":"2024-02-24T08:01:32Z","abstract_excerpt":"Large language models (LLMs) have emerged as a promising alternative to expensive human evaluations. However, the alignment and coverage of LLM-based evaluations are often limited by the scope and potential bias of the evaluation prompts and criteria. To address this challenge, we propose HD-Eval, a novel framework that iteratively aligns LLM-based evaluators with human preference via Hierarchical Criteria Decomposition. HD-Eval inherits the essence from the evaluation mindset of human experts and enhances the alignment of LLM-based evaluators by decomposing a given evaluation task into finer-"},"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":"2402.15754","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-24T08:01:32Z","cross_cats_sorted":[],"title_canon_sha256":"fa7173e2c13c5dc8af351427183b02869e542f1f4aa77c09813ed4ed84fc2c28","abstract_canon_sha256":"f8dc9b2dfaf5c5c02e848e5bd1fb836dba3be396bb126cfb20476a5e27b1113a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:57.619812Z","signature_b64":"eCt9AeHM8H5SlmlhuQIca+yr9FIvyso76JK8R6QVc2BK87V+ESY1WnT8JvQtpR30I/y1Z1pEn2bYrXDmadV3Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7a8a6ed5c79d3ab4971d766fac1c48e9fbc77df692e9477fe1a902dc444c773","last_reissued_at":"2026-07-05T07:48:57.619376Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:57.619376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HD-Eval: Aligning Large Language Model Evaluators Through Hierarchical Criteria Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, Qi Zhang, Shaohan Huang, Tianchi Yang, Weiwei Deng, Yuxuan Liu, Zihan Zhang","submitted_at":"2024-02-24T08:01:32Z","abstract_excerpt":"Large language models (LLMs) have emerged as a promising alternative to expensive human evaluations. However, the alignment and coverage of LLM-based evaluations are often limited by the scope and potential bias of the evaluation prompts and criteria. To address this challenge, we propose HD-Eval, a novel framework that iteratively aligns LLM-based evaluators with human preference via Hierarchical Criteria Decomposition. HD-Eval inherits the essence from the evaluation mindset of human experts and enhances the alignment of LLM-based evaluators by decomposing a given evaluation task into finer-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.15754","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/2402.15754/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":"2402.15754","created_at":"2026-07-05T07:48:57.619435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.15754v1","created_at":"2026-07-05T07:48:57.619435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.15754","created_at":"2026-07-05T07:48:57.619435+00:00"},{"alias_kind":"pith_short_12","alias_value":"U6UKN3K4PHJ2","created_at":"2026-07-05T07:48:57.619435+00:00"},{"alias_kind":"pith_short_16","alias_value":"U6UKN3K4PHJ2WSLR","created_at":"2026-07-05T07:48:57.619435+00:00"},{"alias_kind":"pith_short_8","alias_value":"U6UKN3K4","created_at":"2026-07-05T07:48:57.619435+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07936","citing_title":"Illusions of the Gold Standard: A Large-scale Analysis of Human Evaluation Protocols for Long-form Text Generation","ref_index":92,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16615","citing_title":"Learning What Evaluators Value: A Reliable Approach to Modeling Evaluator Preferences","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16615","citing_title":"Learning What Evaluators Value: A Reliable Approach to Modeling Evaluator Preferences","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2601.03013","citing_title":"LLMs, You Can Evaluate It! Design of Multi-perspective Report Evaluation for Security Operation Centers","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2412.05579","citing_title":"LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods","ref_index":152,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2","json":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2.json","graph_json":"https://pith.science/api/pith-number/U6UKN3K4PHJ2WSLR25TPVQOER2/graph.json","events_json":"https://pith.science/api/pith-number/U6UKN3K4PHJ2WSLR25TPVQOER2/events.json","paper":"https://pith.science/paper/U6UKN3K4"},"agent_actions":{"view_html":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2","download_json":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2.json","view_paper":"https://pith.science/paper/U6UKN3K4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.15754&json=true","fetch_graph":"https://pith.science/api/pith-number/U6UKN3K4PHJ2WSLR25TPVQOER2/graph.json","fetch_events":"https://pith.science/api/pith-number/U6UKN3K4PHJ2WSLR25TPVQOER2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2/action/storage_attestation","attest_author":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2/action/author_attestation","sign_citation":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2/action/citation_signature","submit_replication":"https://pith.science/pith/U6UKN3K4PHJ2WSLR25TPVQOER2/action/replication_record"}},"created_at":"2026-07-05T07:48:57.619435+00:00","updated_at":"2026-07-05T07:48:57.619435+00:00"}