{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4BN74IVUJEJEVGJBNSVAFC5WAD","short_pith_number":"pith:4BN74IVU","schema_version":"1.0","canonical_sha256":"e05bfe22b449124a99216caa028bb600d69cf5fa6066c5dddc460183560e8fa9","source":{"kind":"arxiv","id":"2502.00691","version":4},"attestation_state":"computed","paper":{"title":"To Code or not to Code? Adaptive Tool Integration for Math Language Models via Expectation-Maximization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chao Qu, Fangzhen Lin, Fengming Zhu, Haozhe Wang, Long Li, Wei Chu, Weidi Xu","submitted_at":"2025-02-02T06:32:23Z","abstract_excerpt":"Recent advances in mathematical problem-solving with language models (LMs) integrate chain-of-thought (CoT) reasoning and code execution to harness their complementary strengths. However, existing hybrid frameworks exhibit a critical limitation: they depend on externally dictated instructions or rigid code-integration templates, lacking metacognitive awareness -- the capacity to dynamically evaluate intrinsic capabilities and autonomously determine when and how to integrate tools. This rigidity motivates our study of autonomous code integration, enabling models to adapt tool-usage strategies a"},"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":"2502.00691","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-02T06:32:23Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"a746923a08fbb0d4040678fbe32cc7d76c2d391b6d54ea94acf9033fb2e53d26","abstract_canon_sha256":"95dc71a6173a52849eb0d9b3ff67f919b5e5c3c2974930872d0964bac3fabbcc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:39:07.224965Z","signature_b64":"fbaSZMc+h823+asUzVNGw1+32lBaPuA1vYCIgsEPzHDyR3W4UKHmhUByC72qB3y0I+iIG+0Wcqd8y6AbhttSBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e05bfe22b449124a99216caa028bb600d69cf5fa6066c5dddc460183560e8fa9","last_reissued_at":"2026-07-05T11:39:07.224526Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:39:07.224526Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"To Code or not to Code? Adaptive Tool Integration for Math Language Models via Expectation-Maximization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chao Qu, Fangzhen Lin, Fengming Zhu, Haozhe Wang, Long Li, Wei Chu, Weidi Xu","submitted_at":"2025-02-02T06:32:23Z","abstract_excerpt":"Recent advances in mathematical problem-solving with language models (LMs) integrate chain-of-thought (CoT) reasoning and code execution to harness their complementary strengths. However, existing hybrid frameworks exhibit a critical limitation: they depend on externally dictated instructions or rigid code-integration templates, lacking metacognitive awareness -- the capacity to dynamically evaluate intrinsic capabilities and autonomously determine when and how to integrate tools. This rigidity motivates our study of autonomous code integration, enabling models to adapt tool-usage strategies a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00691","kind":"arxiv","version":4},"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/2502.00691/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":"2502.00691","created_at":"2026-07-05T11:39:07.224578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.00691v4","created_at":"2026-07-05T11:39:07.224578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00691","created_at":"2026-07-05T11:39:07.224578+00:00"},{"alias_kind":"pith_short_12","alias_value":"4BN74IVUJEJE","created_at":"2026-07-05T11:39:07.224578+00:00"},{"alias_kind":"pith_short_16","alias_value":"4BN74IVUJEJEVGJB","created_at":"2026-07-05T11:39:07.224578+00:00"},{"alias_kind":"pith_short_8","alias_value":"4BN74IVU","created_at":"2026-07-05T11:39:07.224578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14054","citing_title":"Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD","json":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD.json","graph_json":"https://pith.science/api/pith-number/4BN74IVUJEJEVGJBNSVAFC5WAD/graph.json","events_json":"https://pith.science/api/pith-number/4BN74IVUJEJEVGJBNSVAFC5WAD/events.json","paper":"https://pith.science/paper/4BN74IVU"},"agent_actions":{"view_html":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD","download_json":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD.json","view_paper":"https://pith.science/paper/4BN74IVU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.00691&json=true","fetch_graph":"https://pith.science/api/pith-number/4BN74IVUJEJEVGJBNSVAFC5WAD/graph.json","fetch_events":"https://pith.science/api/pith-number/4BN74IVUJEJEVGJBNSVAFC5WAD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD/action/storage_attestation","attest_author":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD/action/author_attestation","sign_citation":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD/action/citation_signature","submit_replication":"https://pith.science/pith/4BN74IVUJEJEVGJBNSVAFC5WAD/action/replication_record"}},"created_at":"2026-07-05T11:39:07.224578+00:00","updated_at":"2026-07-05T11:39:07.224578+00:00"}