{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:76M2JRFTXE3EEDQNHP6MSYLNRR","short_pith_number":"pith:76M2JRFT","schema_version":"1.0","canonical_sha256":"ff99a4c4b3b936420e0d3bfcc9616d8c4d1c62d8812f3a22c2115e0e6e060392","source":{"kind":"arxiv","id":"2501.02448","version":2},"attestation_state":"computed","paper":{"title":"Understand, Solve and Translate: Bridging the Multilingual Mathematical Reasoning Gap","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dasol Choi, Guijin Son, Hyunwoo Ko","submitted_at":"2025-01-05T05:57:22Z","abstract_excerpt":"Large language models (LLMs) demonstrate exceptional performance on complex reasoning tasks. However, despite their strong reasoning capabilities in high-resource languages (e.g., English and Chinese), a significant performance gap persists in other languages. To investigate this gap in Korean, we introduce HRM8K, a benchmark comprising 8,011 English-Korean parallel bilingual math problems. Through systematic analysis of model behaviors, we identify a key finding: these performance disparities stem primarily from difficulties in comprehending non-English inputs, rather than limitations in reas"},"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":"2501.02448","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-05T05:57:22Z","cross_cats_sorted":[],"title_canon_sha256":"3f68220e92d20c1c893eb52d71dfbebf120bba8708ed2a21d4873402e4a68206","abstract_canon_sha256":"4c62e97568e7271407b641c504aee9036b2f6a288b006b8936bb01d3cd95f17b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:47.480610Z","signature_b64":"IVELJ3XO7Lus6IM5xHjiUyyvFsH9cVYbnlYkT+IqXiCa3yYDN5Dt4zRe+ppNkB3nBzJgZU5u+Z7OxvSQ1NO8Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff99a4c4b3b936420e0d3bfcc9616d8c4d1c62d8812f3a22c2115e0e6e060392","last_reissued_at":"2026-07-05T10:07:47.480103Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:47.480103Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understand, Solve and Translate: Bridging the Multilingual Mathematical Reasoning Gap","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dasol Choi, Guijin Son, Hyunwoo Ko","submitted_at":"2025-01-05T05:57:22Z","abstract_excerpt":"Large language models (LLMs) demonstrate exceptional performance on complex reasoning tasks. However, despite their strong reasoning capabilities in high-resource languages (e.g., English and Chinese), a significant performance gap persists in other languages. To investigate this gap in Korean, we introduce HRM8K, a benchmark comprising 8,011 English-Korean parallel bilingual math problems. Through systematic analysis of model behaviors, we identify a key finding: these performance disparities stem primarily from difficulties in comprehending non-English inputs, rather than limitations in reas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02448","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/2501.02448/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":"2501.02448","created_at":"2026-07-05T10:07:47.480157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.02448v2","created_at":"2026-07-05T10:07:47.480157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02448","created_at":"2026-07-05T10:07:47.480157+00:00"},{"alias_kind":"pith_short_12","alias_value":"76M2JRFTXE3E","created_at":"2026-07-05T10:07:47.480157+00:00"},{"alias_kind":"pith_short_16","alias_value":"76M2JRFTXE3EEDQN","created_at":"2026-07-05T10:07:47.480157+00:00"},{"alias_kind":"pith_short_8","alias_value":"76M2JRFT","created_at":"2026-07-05T10:07:47.480157+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18453","citing_title":"LLM Parameters for Math Across Languages: Shared or Separate?","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08728","citing_title":"Artificial Intelligence for Mathematical Reasoning: An Integrated Survey of Language Models, Neuro-symbolic Systems, and Verified Discovery","ref_index":224,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07020","citing_title":"MADE: Beyond Scoring via a Multilingual Agentic Diagnosing Engine for Fine-Grained Evaluation Insights","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR","json":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR.json","graph_json":"https://pith.science/api/pith-number/76M2JRFTXE3EEDQNHP6MSYLNRR/graph.json","events_json":"https://pith.science/api/pith-number/76M2JRFTXE3EEDQNHP6MSYLNRR/events.json","paper":"https://pith.science/paper/76M2JRFT"},"agent_actions":{"view_html":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR","download_json":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR.json","view_paper":"https://pith.science/paper/76M2JRFT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.02448&json=true","fetch_graph":"https://pith.science/api/pith-number/76M2JRFTXE3EEDQNHP6MSYLNRR/graph.json","fetch_events":"https://pith.science/api/pith-number/76M2JRFTXE3EEDQNHP6MSYLNRR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR/action/storage_attestation","attest_author":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR/action/author_attestation","sign_citation":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR/action/citation_signature","submit_replication":"https://pith.science/pith/76M2JRFTXE3EEDQNHP6MSYLNRR/action/replication_record"}},"created_at":"2026-07-05T10:07:47.480157+00:00","updated_at":"2026-07-05T10:07:47.480157+00:00"}