{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4T77SKZRVCKP2EVEC76S7HYYAE","short_pith_number":"pith:4T77SKZR","schema_version":"1.0","canonical_sha256":"e4fff92b31a894fd12a417fd2f9f18013ea1734b3a424117d03bdc0ada24feb4","source":{"kind":"arxiv","id":"2605.18827","version":1},"attestation_state":"computed","paper":{"title":"Code-Guided Reasoning for Small Language Models: Evaluating Executable MCQA Scaffolds","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.IR","authors_text":"Amit Sheth, Dhaval Patel, Prateek Biswas, Shuxin Lin, Vedant Khandelwal","submitted_at":"2026-05-12T20:20:06Z","abstract_excerpt":"Multiple-choice QA benchmarks usually evaluate small language models (SLMs) as direct answerers, but deployed language-model systems increasingly rely on external scaffolds such as tools, code, and repeated model calls. We introduce Code-Guided Reasoning (CGR), an evaluation protocol and generated-program resource for measuring when executable reasoning scaffolds improve SLM performance on MCQA tasks. CGR standardizes six components: a normalized item interface, a direct solver prompt, a generator prompt, a Python scaffold, solver-call and extraction helpers, and a three-channel result record."},"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":"2605.18827","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-12T20:20:06Z","cross_cats_sorted":["cs.LG","cs.PL"],"title_canon_sha256":"2ab67d0d8c79962de8c0449ebedb207d346d8e0bffb37fda5e159eecbbdc3143","abstract_canon_sha256":"7453c72796832d35974c3b093fdfc1ab72922baaddae7eccc5b2cb2a90a73c9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:24.676644Z","signature_b64":"lMoV/9dPEOkzq2Y4ye6DQwlCbafZOrmhXFEC3bhDKdZ9zAZd59sVd2TPXp7cchK6MG2QDWnxxg8swAUXSzE7Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4fff92b31a894fd12a417fd2f9f18013ea1734b3a424117d03bdc0ada24feb4","last_reissued_at":"2026-05-20T00:06:24.676053Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:24.676053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Code-Guided Reasoning for Small Language Models: Evaluating Executable MCQA Scaffolds","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.IR","authors_text":"Amit Sheth, Dhaval Patel, Prateek Biswas, Shuxin Lin, Vedant Khandelwal","submitted_at":"2026-05-12T20:20:06Z","abstract_excerpt":"Multiple-choice QA benchmarks usually evaluate small language models (SLMs) as direct answerers, but deployed language-model systems increasingly rely on external scaffolds such as tools, code, and repeated model calls. We introduce Code-Guided Reasoning (CGR), an evaluation protocol and generated-program resource for measuring when executable reasoning scaffolds improve SLM performance on MCQA tasks. CGR standardizes six components: a normalized item interface, a direct solver prompt, a generator prompt, a Python scaffold, solver-call and extraction helpers, and a three-channel result record."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18827","kind":"arxiv","version":1},"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/2605.18827/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":"2605.18827","created_at":"2026-05-20T00:06:24.676150+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18827v1","created_at":"2026-05-20T00:06:24.676150+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18827","created_at":"2026-05-20T00:06:24.676150+00:00"},{"alias_kind":"pith_short_12","alias_value":"4T77SKZRVCKP","created_at":"2026-05-20T00:06:24.676150+00:00"},{"alias_kind":"pith_short_16","alias_value":"4T77SKZRVCKP2EVE","created_at":"2026-05-20T00:06:24.676150+00:00"},{"alias_kind":"pith_short_8","alias_value":"4T77SKZR","created_at":"2026-05-20T00:06:24.676150+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE","json":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE.json","graph_json":"https://pith.science/api/pith-number/4T77SKZRVCKP2EVEC76S7HYYAE/graph.json","events_json":"https://pith.science/api/pith-number/4T77SKZRVCKP2EVEC76S7HYYAE/events.json","paper":"https://pith.science/paper/4T77SKZR"},"agent_actions":{"view_html":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE","download_json":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE.json","view_paper":"https://pith.science/paper/4T77SKZR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18827&json=true","fetch_graph":"https://pith.science/api/pith-number/4T77SKZRVCKP2EVEC76S7HYYAE/graph.json","fetch_events":"https://pith.science/api/pith-number/4T77SKZRVCKP2EVEC76S7HYYAE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE/action/storage_attestation","attest_author":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE/action/author_attestation","sign_citation":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE/action/citation_signature","submit_replication":"https://pith.science/pith/4T77SKZRVCKP2EVEC76S7HYYAE/action/replication_record"}},"created_at":"2026-05-20T00:06:24.676150+00:00","updated_at":"2026-05-20T00:06:24.676150+00:00"}