{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7WGTVI7KI2OV4VWS3RCRJ2S5MM","short_pith_number":"pith:7WGTVI7K","schema_version":"1.0","canonical_sha256":"fd8d3aa3ea469d5e56d2dc4514ea5d6322cfd6d30c941e2f92d4f425f0ce0d21","source":{"kind":"arxiv","id":"2406.06565","version":2},"attestation_state":"computed","paper":{"title":"MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixtures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fuzhao Xue, Graham Neubig, Jinjie Ni, Kabir Jain, Mahir Shah, Xiang Yue, Yang You, Yuntian Deng","submitted_at":"2024-06-03T05:47:05Z","abstract_excerpt":"Evaluating large language models (LLMs) is challenging. Traditional ground-truth-based benchmarks fail to capture the comprehensiveness and nuance of real-world queries, while LLM-as-judge benchmarks suffer from grading biases and limited query quantity. Both of them may also become contaminated over time. User-facing evaluation, such as Chatbot Arena, provides reliable signals but is costly and slow. In this work, we propose MixEval, a new paradigm for establishing efficient, gold-standard LLM evaluation by strategically mixing off-the-shelf benchmarks. It bridges (1) comprehensive and well-d"},"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":"2406.06565","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-03T05:47:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"ef8b4cd68fafff8f985444cec3acf284f74fbee809a363a8b86abae385895776","abstract_canon_sha256":"39af888dd9b134d62b907a27ca5ec1e27ab8e06546be80072dd016f8bcbb9a45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:19:29.009017Z","signature_b64":"vPtRgqPXiRZpKY1E8lYl0KhfzRWv8sBYapZDvvKqbTDbCjYUiRMoaS9wjByKO+a9dewtftkV+aXWKv/Gml3PBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fd8d3aa3ea469d5e56d2dc4514ea5d6322cfd6d30c941e2f92d4f425f0ce0d21","last_reissued_at":"2026-07-05T09:19:29.008532Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:19:29.008532Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixtures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fuzhao Xue, Graham Neubig, Jinjie Ni, Kabir Jain, Mahir Shah, Xiang Yue, Yang You, Yuntian Deng","submitted_at":"2024-06-03T05:47:05Z","abstract_excerpt":"Evaluating large language models (LLMs) is challenging. Traditional ground-truth-based benchmarks fail to capture the comprehensiveness and nuance of real-world queries, while LLM-as-judge benchmarks suffer from grading biases and limited query quantity. Both of them may also become contaminated over time. User-facing evaluation, such as Chatbot Arena, provides reliable signals but is costly and slow. In this work, we propose MixEval, a new paradigm for establishing efficient, gold-standard LLM evaluation by strategically mixing off-the-shelf benchmarks. It bridges (1) comprehensive and well-d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06565","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/2406.06565/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":"2406.06565","created_at":"2026-07-05T09:19:29.008590+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.06565v2","created_at":"2026-07-05T09:19:29.008590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06565","created_at":"2026-07-05T09:19:29.008590+00:00"},{"alias_kind":"pith_short_12","alias_value":"7WGTVI7KI2OV","created_at":"2026-07-05T09:19:29.008590+00:00"},{"alias_kind":"pith_short_16","alias_value":"7WGTVI7KI2OV4VWS","created_at":"2026-07-05T09:19:29.008590+00:00"},{"alias_kind":"pith_short_8","alias_value":"7WGTVI7K","created_at":"2026-07-05T09:19:29.008590+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31268","citing_title":"Mellum2 Technical Report","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM","json":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM.json","graph_json":"https://pith.science/api/pith-number/7WGTVI7KI2OV4VWS3RCRJ2S5MM/graph.json","events_json":"https://pith.science/api/pith-number/7WGTVI7KI2OV4VWS3RCRJ2S5MM/events.json","paper":"https://pith.science/paper/7WGTVI7K"},"agent_actions":{"view_html":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM","download_json":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM.json","view_paper":"https://pith.science/paper/7WGTVI7K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.06565&json=true","fetch_graph":"https://pith.science/api/pith-number/7WGTVI7KI2OV4VWS3RCRJ2S5MM/graph.json","fetch_events":"https://pith.science/api/pith-number/7WGTVI7KI2OV4VWS3RCRJ2S5MM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM/action/storage_attestation","attest_author":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM/action/author_attestation","sign_citation":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM/action/citation_signature","submit_replication":"https://pith.science/pith/7WGTVI7KI2OV4VWS3RCRJ2S5MM/action/replication_record"}},"created_at":"2026-07-05T09:19:29.008590+00:00","updated_at":"2026-07-05T09:19:29.008590+00:00"}