{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RW2PZMXHSSC77INE7LHEJRJFJI","short_pith_number":"pith:RW2PZMXH","schema_version":"1.0","canonical_sha256":"8db4fcb2e79485ffa1a4face44c5254a20fce3c4fdd841906775c8219731c0de","source":{"kind":"arxiv","id":"2407.03624","version":2},"attestation_state":"computed","paper":{"title":"Question-Analysis Prompting Improves LLM Performance in Reasoning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dharunish Yugeswardeenoo, Kevin Zhu, Sean O'Brien","submitted_at":"2024-07-04T04:19:50Z","abstract_excerpt":"Although LLMs have the potential to transform many fields, they still underperform humans in reasoning tasks. Existing methods induce the model to produce step-by-step calculations, but this research explores the question: Does making the LLM analyze the question improve its performance? We propose a novel prompting strategy called Question Analysis Prompting (QAP), in which the model is prompted to explain the question in $n$ words before solving. The value of $n$ influences the length of response generated by the model. QAP is evaluated on GPT 3.5 Turbo and GPT 4 Turbo on arithmetic datasets"},"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":"2407.03624","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-04T04:19:50Z","cross_cats_sorted":[],"title_canon_sha256":"db2332a81570582e7baf0f78fdd5945fa633a56d5fbe1d1defaeb47e12383029","abstract_canon_sha256":"9e5d742b10dbf1ed07a7f972b78ed8540c3818a7ba38fa7a49385ab7ce7b216c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:59:10.532400Z","signature_b64":"nAISCMOU4v1PDpdPLNM3+3+8XmOTStVI1omxYoWyby6zdPwckPlNZiRvdbT4EBvPzh3F8VtCIwKBD7Ro0qa6Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8db4fcb2e79485ffa1a4face44c5254a20fce3c4fdd841906775c8219731c0de","last_reissued_at":"2026-07-05T08:59:10.531850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:59:10.531850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Question-Analysis Prompting Improves LLM Performance in Reasoning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dharunish Yugeswardeenoo, Kevin Zhu, Sean O'Brien","submitted_at":"2024-07-04T04:19:50Z","abstract_excerpt":"Although LLMs have the potential to transform many fields, they still underperform humans in reasoning tasks. Existing methods induce the model to produce step-by-step calculations, but this research explores the question: Does making the LLM analyze the question improve its performance? We propose a novel prompting strategy called Question Analysis Prompting (QAP), in which the model is prompted to explain the question in $n$ words before solving. The value of $n$ influences the length of response generated by the model. QAP is evaluated on GPT 3.5 Turbo and GPT 4 Turbo on arithmetic datasets"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.03624","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/2407.03624/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":"2407.03624","created_at":"2026-07-05T08:59:10.531906+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.03624v2","created_at":"2026-07-05T08:59:10.531906+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.03624","created_at":"2026-07-05T08:59:10.531906+00:00"},{"alias_kind":"pith_short_12","alias_value":"RW2PZMXHSSC7","created_at":"2026-07-05T08:59:10.531906+00:00"},{"alias_kind":"pith_short_16","alias_value":"RW2PZMXHSSC77INE","created_at":"2026-07-05T08:59:10.531906+00:00"},{"alias_kind":"pith_short_8","alias_value":"RW2PZMXH","created_at":"2026-07-05T08:59:10.531906+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.17056","citing_title":"From Test-taking to Cognitive Scaffolding: A Pedagogical Diagnostic Benchmark for LLMs on English Standardized Tests","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI","json":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI.json","graph_json":"https://pith.science/api/pith-number/RW2PZMXHSSC77INE7LHEJRJFJI/graph.json","events_json":"https://pith.science/api/pith-number/RW2PZMXHSSC77INE7LHEJRJFJI/events.json","paper":"https://pith.science/paper/RW2PZMXH"},"agent_actions":{"view_html":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI","download_json":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI.json","view_paper":"https://pith.science/paper/RW2PZMXH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.03624&json=true","fetch_graph":"https://pith.science/api/pith-number/RW2PZMXHSSC77INE7LHEJRJFJI/graph.json","fetch_events":"https://pith.science/api/pith-number/RW2PZMXHSSC77INE7LHEJRJFJI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI/action/storage_attestation","attest_author":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI/action/author_attestation","sign_citation":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI/action/citation_signature","submit_replication":"https://pith.science/pith/RW2PZMXHSSC77INE7LHEJRJFJI/action/replication_record"}},"created_at":"2026-07-05T08:59:10.531906+00:00","updated_at":"2026-07-05T08:59:10.531906+00:00"}