{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WKD3OVPKYI3KN5QAG2TOYCLTLI","short_pith_number":"pith:WKD3OVPK","schema_version":"1.0","canonical_sha256":"b287b755eac236a6f60036a6ec09735a31639d64c319c45de22b268f1872c63b","source":{"kind":"arxiv","id":"2505.14279","version":2},"attestation_state":"computed","paper":{"title":"YESciEval: Robust LLM-as-a-Judge for Scientific Question Answering","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hamed Babaei Giglou, Jennifer D'Souza, Quentin M\\\"unch","submitted_at":"2025-05-20T12:30:46Z","abstract_excerpt":"Large Language Models (LLMs) drive scientific question-answering on modern search engines, yet their evaluation robustness remains underexplored. We introduce YESciEval, an open-source framework that combines fine-grained rubric-based assessment with reinforcement learning to mitigate optimism bias in LLM evaluators. We release multidisciplinary scienceQ&A datasets, including adversarial variants, with evaluation scores from multiple LLMs. Independent of proprietary models and human feedback, our approach enables scalable, cost-free evaluation. By advancing reliable LLM-as-a-judge models, this"},"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":"2505.14279","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T12:30:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"efb3b0d93362ef21e89c58efd8c128b04ced3dc5a18e8de7bfa59173add5d72a","abstract_canon_sha256":"9a33d222920c4d7dbc2c73608fb782409216e1ceb75941c96c7f18c4e0ab0100"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:45.799382Z","signature_b64":"69rGrI/TUGssDjFAlrBfMMWzhLqFGGvQAi3pwpL5HqVga/X6XBmf/U5P1+XCs4l48n/mSfEJ6p4BTph+cDYLAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b287b755eac236a6f60036a6ec09735a31639d64c319c45de22b268f1872c63b","last_reissued_at":"2026-07-05T11:11:45.798904Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:45.798904Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"YESciEval: Robust LLM-as-a-Judge for Scientific Question Answering","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hamed Babaei Giglou, Jennifer D'Souza, Quentin M\\\"unch","submitted_at":"2025-05-20T12:30:46Z","abstract_excerpt":"Large Language Models (LLMs) drive scientific question-answering on modern search engines, yet their evaluation robustness remains underexplored. We introduce YESciEval, an open-source framework that combines fine-grained rubric-based assessment with reinforcement learning to mitigate optimism bias in LLM evaluators. We release multidisciplinary scienceQ&A datasets, including adversarial variants, with evaluation scores from multiple LLMs. Independent of proprietary models and human feedback, our approach enables scalable, cost-free evaluation. By advancing reliable LLM-as-a-judge models, this"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14279","kind":"arxiv","version":2},"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/2505.14279/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":"2505.14279","created_at":"2026-07-05T11:11:45.798957+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14279v2","created_at":"2026-07-05T11:11:45.798957+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14279","created_at":"2026-07-05T11:11:45.798957+00:00"},{"alias_kind":"pith_short_12","alias_value":"WKD3OVPKYI3K","created_at":"2026-07-05T11:11:45.798957+00:00"},{"alias_kind":"pith_short_16","alias_value":"WKD3OVPKYI3KN5QA","created_at":"2026-07-05T11:11:45.798957+00:00"},{"alias_kind":"pith_short_8","alias_value":"WKD3OVPK","created_at":"2026-07-05T11:11:45.798957+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.10522","citing_title":"DeepResearch$^{\\text{Eco}}$: A Recursive Agentic Workflow for Complex Scientific Question Answering in Ecology","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI","json":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI.json","graph_json":"https://pith.science/api/pith-number/WKD3OVPKYI3KN5QAG2TOYCLTLI/graph.json","events_json":"https://pith.science/api/pith-number/WKD3OVPKYI3KN5QAG2TOYCLTLI/events.json","paper":"https://pith.science/paper/WKD3OVPK"},"agent_actions":{"view_html":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI","download_json":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI.json","view_paper":"https://pith.science/paper/WKD3OVPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14279&json=true","fetch_graph":"https://pith.science/api/pith-number/WKD3OVPKYI3KN5QAG2TOYCLTLI/graph.json","fetch_events":"https://pith.science/api/pith-number/WKD3OVPKYI3KN5QAG2TOYCLTLI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI/action/storage_attestation","attest_author":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI/action/author_attestation","sign_citation":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI/action/citation_signature","submit_replication":"https://pith.science/pith/WKD3OVPKYI3KN5QAG2TOYCLTLI/action/replication_record"}},"created_at":"2026-07-05T11:11:45.798957+00:00","updated_at":"2026-07-05T11:11:45.798957+00:00"}