{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7KMGZOHPLKDR75V3Q4NIA75MPO","short_pith_number":"pith:7KMGZOHP","schema_version":"1.0","canonical_sha256":"fa986cb8ef5a871ff6bb871a807fac7ba7bebb2e4c45bab832e7ba3f7fb6921a","source":{"kind":"arxiv","id":"2406.15053","version":2},"attestation_state":"computed","paper":{"title":"PARIKSHA: A Large-Scale Investigation of Human-LLM Evaluator Agreement on Multilingual and Multi-Cultural Data","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aditya Yadavalli, Ishaan Watts, Manohar Swaminathan, Sunayana Sitaram, Varun Gumma, Vivek Seshadri","submitted_at":"2024-06-21T11:00:38Z","abstract_excerpt":"Evaluation of multilingual Large Language Models (LLMs) is challenging due to a variety of factors -- the lack of benchmarks with sufficient linguistic diversity, contamination of popular benchmarks into LLM pre-training data and the lack of local, cultural nuances in translated benchmarks. In this work, we study human and LLM-based evaluation in a multilingual, multi-cultural setting. We evaluate 30 models across 10 Indic languages by conducting 90K human evaluations and 30K LLM-based evaluations and find that models such as GPT-4o and Llama-3 70B consistently perform best for most Indic lang"},"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":"2406.15053","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-21T11:00:38Z","cross_cats_sorted":[],"title_canon_sha256":"6eaa4e38528abbbe348d78d58346893f74d3548685484478677a4689007ba918","abstract_canon_sha256":"045f54cb83edd1e16149a0a4e9f2e8239908851504d6211c0e30dc1081034584"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:22:19.924097Z","signature_b64":"oSn20D6lakY+LETlc0GqjmzpDmc1+wOVTTzDW5iOgx1++nBjsXf8h7NwYndSU+AXMNCzhfV0PdrbJL+7F7KtCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa986cb8ef5a871ff6bb871a807fac7ba7bebb2e4c45bab832e7ba3f7fb6921a","last_reissued_at":"2026-07-05T09:22:19.923509Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:22:19.923509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PARIKSHA: A Large-Scale Investigation of Human-LLM Evaluator Agreement on Multilingual and Multi-Cultural Data","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aditya Yadavalli, Ishaan Watts, Manohar Swaminathan, Sunayana Sitaram, Varun Gumma, Vivek Seshadri","submitted_at":"2024-06-21T11:00:38Z","abstract_excerpt":"Evaluation of multilingual Large Language Models (LLMs) is challenging due to a variety of factors -- the lack of benchmarks with sufficient linguistic diversity, contamination of popular benchmarks into LLM pre-training data and the lack of local, cultural nuances in translated benchmarks. In this work, we study human and LLM-based evaluation in a multilingual, multi-cultural setting. We evaluate 30 models across 10 Indic languages by conducting 90K human evaluations and 30K LLM-based evaluations and find that models such as GPT-4o and Llama-3 70B consistently perform best for most Indic lang"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15053","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/2406.15053/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":"2406.15053","created_at":"2026-07-05T09:22:19.923570+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.15053v2","created_at":"2026-07-05T09:22:19.923570+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15053","created_at":"2026-07-05T09:22:19.923570+00:00"},{"alias_kind":"pith_short_12","alias_value":"7KMGZOHPLKDR","created_at":"2026-07-05T09:22:19.923570+00:00"},{"alias_kind":"pith_short_16","alias_value":"7KMGZOHPLKDR75V3","created_at":"2026-07-05T09:22:19.923570+00:00"},{"alias_kind":"pith_short_8","alias_value":"7KMGZOHP","created_at":"2026-07-05T09:22:19.923570+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.05579","citing_title":"LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods","ref_index":250,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO","json":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO.json","graph_json":"https://pith.science/api/pith-number/7KMGZOHPLKDR75V3Q4NIA75MPO/graph.json","events_json":"https://pith.science/api/pith-number/7KMGZOHPLKDR75V3Q4NIA75MPO/events.json","paper":"https://pith.science/paper/7KMGZOHP"},"agent_actions":{"view_html":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO","download_json":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO.json","view_paper":"https://pith.science/paper/7KMGZOHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.15053&json=true","fetch_graph":"https://pith.science/api/pith-number/7KMGZOHPLKDR75V3Q4NIA75MPO/graph.json","fetch_events":"https://pith.science/api/pith-number/7KMGZOHPLKDR75V3Q4NIA75MPO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO/action/storage_attestation","attest_author":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO/action/author_attestation","sign_citation":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO/action/citation_signature","submit_replication":"https://pith.science/pith/7KMGZOHPLKDR75V3Q4NIA75MPO/action/replication_record"}},"created_at":"2026-07-05T09:22:19.923570+00:00","updated_at":"2026-07-05T09:22:19.923570+00:00"}