{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3PSWQBCZPFPDMNPVQVIVJQKCYI","short_pith_number":"pith:3PSWQBCZ","schema_version":"1.0","canonical_sha256":"dbe5680459795e3635f5855154c142c22827b71a4e24f85de76e13bb9d38e610","source":{"kind":"arxiv","id":"2410.01504","version":2},"attestation_state":"computed","paper":{"title":"PersonaMath: Boosting Mathematical Reasoning via Persona-Driven Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ahmadreza Argha, Chang Ao, Chengming Li, Hamid Alinejad-Rokny, Jiaming Li, Jiayuan Su, Jing Luo, Liang Zhu, Longze Chen, Min Yang, Run Luo, Shiwen Ni, Wen Yang, Xin Cheng, Yukun Chen","submitted_at":"2024-10-02T12:57:12Z","abstract_excerpt":"While closed-source Large Language Models (LLMs) demonstrate strong mathematical problem-solving abilities, open-source models still face challenges with such tasks. To bridge this gap, we propose a data augmentation approach and introduce PersonaMathQA, a dataset derived from MATH and GSM8K, on which we train the PersonaMath models. Our approach consists of two stages: the first stage focuses on learning from Persona Diversification, and the second stage emphasizes learning from Reflection. In the first stage, we regenerate detailed chain-of-thought (CoT) solutions as instructions using a clo"},"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":"2410.01504","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-02T12:57:12Z","cross_cats_sorted":[],"title_canon_sha256":"c5cb52e7dbf81517aac378027a232d749c2c0e36090db4e4b81396e146a60599","abstract_canon_sha256":"94f6ec531d9482264aa0662ada403856d4bb6fa1df904c664fd545cee6801187"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:50.056485Z","signature_b64":"2qO7UfL2HW3k0ikVGkfKl8pkSNG3bZKes5r8IJEw6mJKiF2lszLQY1+QmAxQWxN5PWAqA1QxkxsrRKJ9tzM6Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dbe5680459795e3635f5855154c142c22827b71a4e24f85de76e13bb9d38e610","last_reissued_at":"2026-07-05T10:17:50.055976Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:50.055976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PersonaMath: Boosting Mathematical Reasoning via Persona-Driven Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ahmadreza Argha, Chang Ao, Chengming Li, Hamid Alinejad-Rokny, Jiaming Li, Jiayuan Su, Jing Luo, Liang Zhu, Longze Chen, Min Yang, Run Luo, Shiwen Ni, Wen Yang, Xin Cheng, Yukun Chen","submitted_at":"2024-10-02T12:57:12Z","abstract_excerpt":"While closed-source Large Language Models (LLMs) demonstrate strong mathematical problem-solving abilities, open-source models still face challenges with such tasks. To bridge this gap, we propose a data augmentation approach and introduce PersonaMathQA, a dataset derived from MATH and GSM8K, on which we train the PersonaMath models. Our approach consists of two stages: the first stage focuses on learning from Persona Diversification, and the second stage emphasizes learning from Reflection. In the first stage, we regenerate detailed chain-of-thought (CoT) solutions as instructions using a clo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01504","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/2410.01504/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":"2410.01504","created_at":"2026-07-05T10:17:50.056035+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01504v2","created_at":"2026-07-05T10:17:50.056035+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01504","created_at":"2026-07-05T10:17:50.056035+00:00"},{"alias_kind":"pith_short_12","alias_value":"3PSWQBCZPFPD","created_at":"2026-07-05T10:17:50.056035+00:00"},{"alias_kind":"pith_short_16","alias_value":"3PSWQBCZPFPDMNPV","created_at":"2026-07-05T10:17:50.056035+00:00"},{"alias_kind":"pith_short_8","alias_value":"3PSWQBCZ","created_at":"2026-07-05T10:17:50.056035+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.24160","citing_title":"Towards Automated Crowdsourced Testing via Personified-LLM","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI","json":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI.json","graph_json":"https://pith.science/api/pith-number/3PSWQBCZPFPDMNPVQVIVJQKCYI/graph.json","events_json":"https://pith.science/api/pith-number/3PSWQBCZPFPDMNPVQVIVJQKCYI/events.json","paper":"https://pith.science/paper/3PSWQBCZ"},"agent_actions":{"view_html":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI","download_json":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI.json","view_paper":"https://pith.science/paper/3PSWQBCZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01504&json=true","fetch_graph":"https://pith.science/api/pith-number/3PSWQBCZPFPDMNPVQVIVJQKCYI/graph.json","fetch_events":"https://pith.science/api/pith-number/3PSWQBCZPFPDMNPVQVIVJQKCYI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI/action/storage_attestation","attest_author":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI/action/author_attestation","sign_citation":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI/action/citation_signature","submit_replication":"https://pith.science/pith/3PSWQBCZPFPDMNPVQVIVJQKCYI/action/replication_record"}},"created_at":"2026-07-05T10:17:50.056035+00:00","updated_at":"2026-07-05T10:17:50.056035+00:00"}