{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:D5QLEIIWEAVWATKUJCPIDB3DTQ","short_pith_number":"pith:D5QLEIIW","schema_version":"1.0","canonical_sha256":"1f60b22116202b604d54489e8187639c379ccef3313d9aa80d836b2ffe453c82","source":{"kind":"arxiv","id":"2405.11966","version":4},"attestation_state":"computed","paper":{"title":"Multiple-Choice Questions are Efficient and Robust LLM Evaluators","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hongkun Hao, Lizhen Xu, Rui Wang, Zhaokun Jiang, Ziyin Zhang","submitted_at":"2024-05-20T11:47:13Z","abstract_excerpt":"We present GSM-MC, a multiple-choice (MC) dataset constructed by collecting answers and incorrect predictions on GSM8K from 60 open-source models. Through extensive experiments, we show that LLMs' performance on the MC version of this popular benchmark is strongly correlated with their performance on the original version and is quite robust to distractor choices and option orders, while the evaluation time is reduced by a factor of up to 30. Following similar procedures, we introduce MATH-MC, constructed from MATH, and PythonIO, a new program reasoning MC dataset constructed from HumanEval and"},"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":"2405.11966","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-20T11:47:13Z","cross_cats_sorted":[],"title_canon_sha256":"c41a304e41bfc984e88e9782fa6af577e3890011de6dc0e51866d350b747dc09","abstract_canon_sha256":"9827efb9b29702f921d7dc05b75b26bcfb6136609634ca291d2c2d223244aa7e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:47.764761Z","signature_b64":"THKSuH9rDIrisAlpuvrIqG3b2SEo6yT/Wl+FTMH4m+w+FrpwiGv3z1/Dg4H1jITLxq6THenv/fhKHsA4newoBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f60b22116202b604d54489e8187639c379ccef3313d9aa80d836b2ffe453c82","last_reissued_at":"2026-07-05T08:36:47.764187Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:47.764187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multiple-Choice Questions are Efficient and Robust LLM Evaluators","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hongkun Hao, Lizhen Xu, Rui Wang, Zhaokun Jiang, Ziyin Zhang","submitted_at":"2024-05-20T11:47:13Z","abstract_excerpt":"We present GSM-MC, a multiple-choice (MC) dataset constructed by collecting answers and incorrect predictions on GSM8K from 60 open-source models. Through extensive experiments, we show that LLMs' performance on the MC version of this popular benchmark is strongly correlated with their performance on the original version and is quite robust to distractor choices and option orders, while the evaluation time is reduced by a factor of up to 30. Following similar procedures, we introduce MATH-MC, constructed from MATH, and PythonIO, a new program reasoning MC dataset constructed from HumanEval and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.11966","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/2405.11966/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":"2405.11966","created_at":"2026-07-05T08:36:47.764257+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.11966v4","created_at":"2026-07-05T08:36:47.764257+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.11966","created_at":"2026-07-05T08:36:47.764257+00:00"},{"alias_kind":"pith_short_12","alias_value":"D5QLEIIWEAVW","created_at":"2026-07-05T08:36:47.764257+00:00"},{"alias_kind":"pith_short_16","alias_value":"D5QLEIIWEAVWATKU","created_at":"2026-07-05T08:36:47.764257+00:00"},{"alias_kind":"pith_short_8","alias_value":"D5QLEIIW","created_at":"2026-07-05T08:36:47.764257+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06648","citing_title":"Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20189","citing_title":"SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation","ref_index":70,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ","json":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ.json","graph_json":"https://pith.science/api/pith-number/D5QLEIIWEAVWATKUJCPIDB3DTQ/graph.json","events_json":"https://pith.science/api/pith-number/D5QLEIIWEAVWATKUJCPIDB3DTQ/events.json","paper":"https://pith.science/paper/D5QLEIIW"},"agent_actions":{"view_html":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ","download_json":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ.json","view_paper":"https://pith.science/paper/D5QLEIIW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.11966&json=true","fetch_graph":"https://pith.science/api/pith-number/D5QLEIIWEAVWATKUJCPIDB3DTQ/graph.json","fetch_events":"https://pith.science/api/pith-number/D5QLEIIWEAVWATKUJCPIDB3DTQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ/action/storage_attestation","attest_author":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ/action/author_attestation","sign_citation":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ/action/citation_signature","submit_replication":"https://pith.science/pith/D5QLEIIWEAVWATKUJCPIDB3DTQ/action/replication_record"}},"created_at":"2026-07-05T08:36:47.764257+00:00","updated_at":"2026-07-05T08:36:47.764257+00:00"}