{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TXAUUGXTYJOAVNWP6DGYIWDHTZ","short_pith_number":"pith:TXAUUGXT","schema_version":"1.0","canonical_sha256":"9dc14a1af3c25c0ab6cff0cd8458679e7bbc13a27ce1709e4e12dea71aee4c90","source":{"kind":"arxiv","id":"2605.16881","version":1},"attestation_state":"computed","paper":{"title":"PaliBench: A Multi-Reference Blueprint for Classical Language Translation Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"PaliBench shows how to build multi-reference benchmarks for classical language translation from existing scholarly translations without treating any one as definitive.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"M\\'at\\'e Metzger, Nadnapang Phophichit","submitted_at":"2026-05-16T08:43:01Z","abstract_excerpt":"Digital humanities projects increasingly rely on machine translation and large language models to widen access to classical, religious, and otherwise under-translated textual traditions. Yet standard translation benchmarks are poorly suited to such materials: they typically compare a system output against a single reference translation, even though classical texts often support multiple faithful renderings that differ in terminology, register, and interpretation. This article introduces PaliBench, both a benchmark for Pali-to-English translation and a reusable method for constructing multi-ref"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.16881","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-16T08:43:01Z","cross_cats_sorted":[],"title_canon_sha256":"3e8846d19b2b787459b172fa9418d9d894a88af412677d7d4baaa904fd1ae6e1","abstract_canon_sha256":"0795e511740c72ea4e04afd40cc13e830b9c9ccf021d99dd85931f7131336435"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:28.007020Z","signature_b64":"Y8oJX38pKqMae6a6/OaiD7a58sxsq0+EW4IwAmK/3C8lSD0r90fXWe5kcNrPZE7po4OJn8Azwgvwf1M3qnDhAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dc14a1af3c25c0ab6cff0cd8458679e7bbc13a27ce1709e4e12dea71aee4c90","last_reissued_at":"2026-05-20T00:03:28.006165Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:28.006165Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PaliBench: A Multi-Reference Blueprint for Classical Language Translation Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"PaliBench shows how to build multi-reference benchmarks for classical language translation from existing scholarly translations without treating any one as definitive.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"M\\'at\\'e Metzger, Nadnapang Phophichit","submitted_at":"2026-05-16T08:43:01Z","abstract_excerpt":"Digital humanities projects increasingly rely on machine translation and large language models to widen access to classical, religious, and otherwise under-translated textual traditions. Yet standard translation benchmarks are poorly suited to such materials: they typically compare a system output against a single reference translation, even though classical texts often support multiple faithful renderings that differ in terminology, register, and interpretation. This article introduces PaliBench, both a benchmark for Pali-to-English translation and a reusable method for constructing multi-ref"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The broader contribution is methodological: PaliBench shows how existing scholarly translations can be transformed into evaluation infrastructure for interpretive textual traditions without treating any single translation as definitive.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that LLM-assisted alignment of independently segmented translations, automated verification, passage-level quality filtering, and deduplication produce a reliable, unbiased multi-reference dataset that faithfully captures interpretive variation across the three human translations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"PaliBench is a multi-reference benchmark for Pali-to-English translation built from 1,700 aligned passages of the Sutta Pitaka using three scholarly translations to evaluate LLMs with complementary metrics.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"PaliBench shows how to build multi-reference benchmarks for classical language translation from existing scholarly translations without treating any one as definitive.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"58a6c0f7a21424a960cfbd7892599353d3251e75083bc22cbd4a62c325eb2266"},"source":{"id":"2605.16881","kind":"arxiv","version":1},"verdict":{"id":"f679905b-abc4-48e4-ac5c-d71639a7d3c4","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:00:33.716397Z","strongest_claim":"The broader contribution is methodological: PaliBench shows how existing scholarly translations can be transformed into evaluation infrastructure for interpretive textual traditions without treating any single translation as definitive.","one_line_summary":"PaliBench is a multi-reference benchmark for Pali-to-English translation built from 1,700 aligned passages of the Sutta Pitaka using three scholarly translations to evaluate LLMs with complementary metrics.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that LLM-assisted alignment of independently segmented translations, automated verification, passage-level quality filtering, and deduplication produce a reliable, unbiased multi-reference dataset that faithfully captures interpretive variation across the three human translations.","pith_extraction_headline":"PaliBench shows how to build multi-reference benchmarks for classical language translation from existing scholarly translations without treating any one as definitive."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16881/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.183659Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:11:37.242047Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.290631Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.368031Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"27727e008c155e263c2cb1bea3eb3dca9234344b9017291595d27597933d4c6c"},"references":{"count":38,"sample":[{"doi":"","year":null,"title":"Frequently Asked Questions About Access to Insight , year =","work_id":"4f19a98b-41ca-46b4-ba08-5fd55274d8ee","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Nature , volume =","work_id":"687d6744-5f52-4fd0-b593-bfce152bc37e","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Bamman, David and Burns, Patrick J. , title =. 2020 , note =","work_id":"aaacc914-b24d-41cf-bac1-19774e5001fd","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Bodhi, Bhikkhu , title =","work_id":"b79a5572-9912-43db-83bb-e35cbc5a950f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"The Middle Length Discourses of the Buddha: A Translation of the Majjhima Nik","work_id":"508e12a3-1de2-43be-9e27-19b166d8e7b9","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":38,"snapshot_sha256":"9cc3e6aadbd7ec74d67e34bae28e00ec98a5e498b2aa2408700f1810e6ddf924","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a397278901d037335e7feeb6e0cb92d53f6b97b85feaac29efa1dbd07ca5fcf6"},"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":"2605.16881","created_at":"2026-05-20T00:03:28.006325+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16881v1","created_at":"2026-05-20T00:03:28.006325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16881","created_at":"2026-05-20T00:03:28.006325+00:00"},{"alias_kind":"pith_short_12","alias_value":"TXAUUGXTYJOA","created_at":"2026-05-20T00:03:28.006325+00:00"},{"alias_kind":"pith_short_16","alias_value":"TXAUUGXTYJOAVNWP","created_at":"2026-05-20T00:03:28.006325+00:00"},{"alias_kind":"pith_short_8","alias_value":"TXAUUGXT","created_at":"2026-05-20T00:03:28.006325+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ","json":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ.json","graph_json":"https://pith.science/api/pith-number/TXAUUGXTYJOAVNWP6DGYIWDHTZ/graph.json","events_json":"https://pith.science/api/pith-number/TXAUUGXTYJOAVNWP6DGYIWDHTZ/events.json","paper":"https://pith.science/paper/TXAUUGXT"},"agent_actions":{"view_html":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ","download_json":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ.json","view_paper":"https://pith.science/paper/TXAUUGXT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16881&json=true","fetch_graph":"https://pith.science/api/pith-number/TXAUUGXTYJOAVNWP6DGYIWDHTZ/graph.json","fetch_events":"https://pith.science/api/pith-number/TXAUUGXTYJOAVNWP6DGYIWDHTZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ/action/storage_attestation","attest_author":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ/action/author_attestation","sign_citation":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ/action/citation_signature","submit_replication":"https://pith.science/pith/TXAUUGXTYJOAVNWP6DGYIWDHTZ/action/replication_record"}},"created_at":"2026-05-20T00:03:28.006325+00:00","updated_at":"2026-05-20T00:03:28.006325+00:00"}