{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IF3AWJJMDG2ZGUJXOKL4ZIWS2A","short_pith_number":"pith:IF3AWJJM","schema_version":"1.0","canonical_sha256":"41760b252c19b59351377297cca2d2d036f9854b8e61915ef520df8dffcd165f","source":{"kind":"arxiv","id":"2502.17956","version":2},"attestation_state":"computed","paper":{"title":"Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Udomcharoenchaikit, Ekapol Chuangsuwanich, Jinheon Baek, Patomporn Payoungkhamdee, Peerat Limkonchotiwat, Potsawee Manakul, Pume Tuchinda, Samuel Cahyawijaya, Sarana Nutanong","submitted_at":"2025-02-25T08:27:28Z","abstract_excerpt":"Multi-step reasoning is essential for large language models (LLMs), yet multilingual performance remains challenging. While Chain-of-Thought (CoT) prompting improves reasoning, it struggles with non-English languages due to the entanglement of reasoning and execution. Program-of-Thought (PoT) prompting separates reasoning from execution, offering a promising alternative but shifting the challenge to generating programs from non-English questions. We propose a framework to evaluate PoT by separating multilingual reasoning from code execution to examine (i) the impact of fine-tuning on question-"},"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":"2502.17956","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T08:27:28Z","cross_cats_sorted":[],"title_canon_sha256":"2af48aed9e7cb14109421d0b4787be68094fa308e06c653516730caec5790e95","abstract_canon_sha256":"07096b0eea30dd5542f66d784682562ca9ebb6779f42c32a2aa78abb7b6355be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:27.955854Z","signature_b64":"ubdlcrM8eS4TcnqgWYy/RerBtXvhSDP0qWAXmyY9vucYZ2i0E8qmoDVGS/7fXpSjiDxNjUI6Fl+bUalETfgDBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41760b252c19b59351377297cca2d2d036f9854b8e61915ef520df8dffcd165f","last_reissued_at":"2026-06-04T01:08:27.955027Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:27.955027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Udomcharoenchaikit, Ekapol Chuangsuwanich, Jinheon Baek, Patomporn Payoungkhamdee, Peerat Limkonchotiwat, Potsawee Manakul, Pume Tuchinda, Samuel Cahyawijaya, Sarana Nutanong","submitted_at":"2025-02-25T08:27:28Z","abstract_excerpt":"Multi-step reasoning is essential for large language models (LLMs), yet multilingual performance remains challenging. While Chain-of-Thought (CoT) prompting improves reasoning, it struggles with non-English languages due to the entanglement of reasoning and execution. Program-of-Thought (PoT) prompting separates reasoning from execution, offering a promising alternative but shifting the challenge to generating programs from non-English questions. We propose a framework to evaluate PoT by separating multilingual reasoning from code execution to examine (i) the impact of fine-tuning on question-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17956","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/2502.17956/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":"2502.17956","created_at":"2026-06-04T01:08:27.955147+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.17956v2","created_at":"2026-06-04T01:08:27.955147+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17956","created_at":"2026-06-04T01:08:27.955147+00:00"},{"alias_kind":"pith_short_12","alias_value":"IF3AWJJMDG2Z","created_at":"2026-06-04T01:08:27.955147+00:00"},{"alias_kind":"pith_short_16","alias_value":"IF3AWJJMDG2ZGUJX","created_at":"2026-06-04T01:08:27.955147+00:00"},{"alias_kind":"pith_short_8","alias_value":"IF3AWJJM","created_at":"2026-06-04T01:08:27.955147+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.02761","citing_title":"Sustainability Analysis of Prompt Strategies for SLM-based Automated Test Generation","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A","json":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A.json","graph_json":"https://pith.science/api/pith-number/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/graph.json","events_json":"https://pith.science/api/pith-number/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/events.json","paper":"https://pith.science/paper/IF3AWJJM"},"agent_actions":{"view_html":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A","download_json":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A.json","view_paper":"https://pith.science/paper/IF3AWJJM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.17956&json=true","fetch_graph":"https://pith.science/api/pith-number/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/graph.json","fetch_events":"https://pith.science/api/pith-number/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/action/storage_attestation","attest_author":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/action/author_attestation","sign_citation":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/action/citation_signature","submit_replication":"https://pith.science/pith/IF3AWJJMDG2ZGUJXOKL4ZIWS2A/action/replication_record"}},"created_at":"2026-06-04T01:08:27.955147+00:00","updated_at":"2026-06-04T01:08:27.955147+00:00"}