{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EBQFGVRISQNDBHTTFDXGONHDYX","short_pith_number":"pith:EBQFGVRI","schema_version":"1.0","canonical_sha256":"2060535628941a309e7328ee6734e3c5f8223af74722524b00de52279c64642b","source":{"kind":"arxiv","id":"2409.06595","version":3},"attestation_state":"computed","paper":{"title":"GroUSE: A Benchmark to Evaluate Evaluators in Grounded Question Answering","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ant\\'onio Loison, Bilel Omrani, Gautier Viaud, Sacha Muller","submitted_at":"2024-09-10T15:39:32Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) has emerged as a common paradigm to use Large Language Models (LLMs) alongside private and up-to-date knowledge bases. In this work, we address the challenges of using LLM-as-a-Judge when evaluating grounded answers generated by RAG systems. To assess the calibration and discrimination capabilities of judge models, we identify 7 generator failure modes and introduce GroUSE (Grounded QA Unitary Scoring of Evaluators), a meta-evaluation benchmark of 144 unit tests. This benchmark reveals that existing automated RAG evaluation frameworks often overlook importa"},"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":"2409.06595","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T15:39:32Z","cross_cats_sorted":[],"title_canon_sha256":"23691a03083ee3b7f6799fec4fe9535ded7284cbe0c7a8fba719ffb65f12d409","abstract_canon_sha256":"f536445f6b482a0a81d4d63e3c50d4c1f92075bd58c40310a813e2a72d49d3d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:27.041327Z","signature_b64":"oaeSu70ivxt3osAApQhZ31X3p7T0+NZOh3NLto1Oj7zldT57hr0DkSI5ThqM7t/2H4QsrHbEmHGXZ7T4KkEzAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2060535628941a309e7328ee6734e3c5f8223af74722524b00de52279c64642b","last_reissued_at":"2026-07-05T10:07:27.040870Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:27.040870Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GroUSE: A Benchmark to Evaluate Evaluators in Grounded Question Answering","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ant\\'onio Loison, Bilel Omrani, Gautier Viaud, Sacha Muller","submitted_at":"2024-09-10T15:39:32Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) has emerged as a common paradigm to use Large Language Models (LLMs) alongside private and up-to-date knowledge bases. In this work, we address the challenges of using LLM-as-a-Judge when evaluating grounded answers generated by RAG systems. To assess the calibration and discrimination capabilities of judge models, we identify 7 generator failure modes and introduce GroUSE (Grounded QA Unitary Scoring of Evaluators), a meta-evaluation benchmark of 144 unit tests. This benchmark reveals that existing automated RAG evaluation frameworks often overlook importa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.06595","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/2409.06595/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":"2409.06595","created_at":"2026-07-05T10:07:27.040927+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.06595v3","created_at":"2026-07-05T10:07:27.040927+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.06595","created_at":"2026-07-05T10:07:27.040927+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBQFGVRISQND","created_at":"2026-07-05T10:07:27.040927+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBQFGVRISQNDBHTT","created_at":"2026-07-05T10:07:27.040927+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBQFGVRI","created_at":"2026-07-05T10:07:27.040927+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01814","citing_title":"MMIR-TCM: Memory-Integrated Multimodal Inference and Retrieval for TCM Clinical Decision Support","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX","json":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX.json","graph_json":"https://pith.science/api/pith-number/EBQFGVRISQNDBHTTFDXGONHDYX/graph.json","events_json":"https://pith.science/api/pith-number/EBQFGVRISQNDBHTTFDXGONHDYX/events.json","paper":"https://pith.science/paper/EBQFGVRI"},"agent_actions":{"view_html":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX","download_json":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX.json","view_paper":"https://pith.science/paper/EBQFGVRI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.06595&json=true","fetch_graph":"https://pith.science/api/pith-number/EBQFGVRISQNDBHTTFDXGONHDYX/graph.json","fetch_events":"https://pith.science/api/pith-number/EBQFGVRISQNDBHTTFDXGONHDYX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX/action/storage_attestation","attest_author":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX/action/author_attestation","sign_citation":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX/action/citation_signature","submit_replication":"https://pith.science/pith/EBQFGVRISQNDBHTTFDXGONHDYX/action/replication_record"}},"created_at":"2026-07-05T10:07:27.040927+00:00","updated_at":"2026-07-05T10:07:27.040927+00:00"}