{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WIAL5INLIKIIEZ24MJNJTW5MD2","short_pith_number":"pith:WIAL5INL","schema_version":"1.0","canonical_sha256":"b200bea1ab429082675c625a99dbac1e860f18f6a60639c52978bac0668a0e9a","source":{"kind":"arxiv","id":"2406.02864","version":1},"attestation_state":"computed","paper":{"title":"NUMCoT: Numerals and Units of Measurement in Chain-of-Thought Reasoning using Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ancheng Xu, Lei Wang, Minghuan Tan, Min Yang, Ruifeng Xu","submitted_at":"2024-06-05T02:26:14Z","abstract_excerpt":"Numeral systems and units of measurement are two conjoined topics in activities of human beings and have mutual effects with the languages expressing them. Currently, the evaluation of Large Language Models (LLMs) often involves mathematical reasoning, yet little attention is given to how minor changes in numbers or units can drastically alter the complexity of problems and the performance of LLMs. In this paper, we scrutinize existing LLMs on processing of numerals and units of measurement by constructing datasets with perturbations. We first anatomize the reasoning of math word problems to d"},"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":"2406.02864","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-05T02:26:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6084e80fb3883805f66e920beb5dd7b875af50d81f76faa2d43d76a35ec6b04a","abstract_canon_sha256":"3c595254d6ab81a60394c641c05ee51177028a1ea7e7ef2627c1021921f31a47"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:31.441519Z","signature_b64":"PcMyUomQQMTfuqLef3BcD6yCDadm+1FWvcVIRepJnQrcMtPQLifmX/f0te5v2MllNeqx8gICcxZGkWfytBsOAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b200bea1ab429082675c625a99dbac1e860f18f6a60639c52978bac0668a0e9a","last_reissued_at":"2026-07-05T08:27:31.441027Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:31.441027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NUMCoT: Numerals and Units of Measurement in Chain-of-Thought Reasoning using Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ancheng Xu, Lei Wang, Minghuan Tan, Min Yang, Ruifeng Xu","submitted_at":"2024-06-05T02:26:14Z","abstract_excerpt":"Numeral systems and units of measurement are two conjoined topics in activities of human beings and have mutual effects with the languages expressing them. Currently, the evaluation of Large Language Models (LLMs) often involves mathematical reasoning, yet little attention is given to how minor changes in numbers or units can drastically alter the complexity of problems and the performance of LLMs. In this paper, we scrutinize existing LLMs on processing of numerals and units of measurement by constructing datasets with perturbations. We first anatomize the reasoning of math word problems to d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02864","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/2406.02864/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":"2406.02864","created_at":"2026-07-05T08:27:31.441084+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.02864v1","created_at":"2026-07-05T08:27:31.441084+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02864","created_at":"2026-07-05T08:27:31.441084+00:00"},{"alias_kind":"pith_short_12","alias_value":"WIAL5INLIKII","created_at":"2026-07-05T08:27:31.441084+00:00"},{"alias_kind":"pith_short_16","alias_value":"WIAL5INLIKIIEZ24","created_at":"2026-07-05T08:27:31.441084+00:00"},{"alias_kind":"pith_short_8","alias_value":"WIAL5INL","created_at":"2026-07-05T08:27:31.441084+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/WIAL5INLIKIIEZ24MJNJTW5MD2","json":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2.json","graph_json":"https://pith.science/api/pith-number/WIAL5INLIKIIEZ24MJNJTW5MD2/graph.json","events_json":"https://pith.science/api/pith-number/WIAL5INLIKIIEZ24MJNJTW5MD2/events.json","paper":"https://pith.science/paper/WIAL5INL"},"agent_actions":{"view_html":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2","download_json":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2.json","view_paper":"https://pith.science/paper/WIAL5INL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.02864&json=true","fetch_graph":"https://pith.science/api/pith-number/WIAL5INLIKIIEZ24MJNJTW5MD2/graph.json","fetch_events":"https://pith.science/api/pith-number/WIAL5INLIKIIEZ24MJNJTW5MD2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2/action/storage_attestation","attest_author":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2/action/author_attestation","sign_citation":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2/action/citation_signature","submit_replication":"https://pith.science/pith/WIAL5INLIKIIEZ24MJNJTW5MD2/action/replication_record"}},"created_at":"2026-07-05T08:27:31.441084+00:00","updated_at":"2026-07-05T08:27:31.441084+00:00"}