{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A2SFIPNYOMYHJVADXEGLFDTADL","short_pith_number":"pith:A2SFIPNY","schema_version":"1.0","canonical_sha256":"06a4543db8733074d403b90cb28e601ad21078a2c75f6583ee35b131e716cbc6","source":{"kind":"arxiv","id":"2505.14178","version":1},"attestation_state":"computed","paper":{"title":"Tokenization Constraints in LLMs: A Study of Symbolic and Arithmetic Reasoning Limits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenyu You, Jiaqi Wei, Juntai Cao, Xiang Zhang, Yiwei Xu","submitted_at":"2025-05-20T10:32:30Z","abstract_excerpt":"Tokenization is the first - and often underappreciated - layer of computation in language models. While Chain-of-Thought (CoT) prompting enables transformer models to approximate recurrent computation by externalizing intermediate steps, we show that the success of such reasoning is fundamentally bounded by the structure of tokenized inputs. This work presents a theoretical and empirical investigation into how tokenization schemes, particularly subword-based methods like byte-pair encoding (BPE), impede symbolic computation by merging or obscuring atomic reasoning units. We introduce the notio"},"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":"2505.14178","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T10:32:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b4b4056f0cbb221ed82ba8ca55ecd380f3344e14b18b7847a787d23a299ebc9c","abstract_canon_sha256":"a21b7ab38a16f958d6bb907647a5e1ab882676edb32b934bafae5ccb859e3d5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:57.250133Z","signature_b64":"5VHqWbMMd8NTk3OzrmBuQHg8PthEQrKMMcfHE6vPuejTOnQEDp79v/sWGy/O2KKO3FOb03OeIwFIAMa2tipbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06a4543db8733074d403b90cb28e601ad21078a2c75f6583ee35b131e716cbc6","last_reissued_at":"2026-07-05T11:05:57.249588Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:57.249588Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tokenization Constraints in LLMs: A Study of Symbolic and Arithmetic Reasoning Limits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenyu You, Jiaqi Wei, Juntai Cao, Xiang Zhang, Yiwei Xu","submitted_at":"2025-05-20T10:32:30Z","abstract_excerpt":"Tokenization is the first - and often underappreciated - layer of computation in language models. While Chain-of-Thought (CoT) prompting enables transformer models to approximate recurrent computation by externalizing intermediate steps, we show that the success of such reasoning is fundamentally bounded by the structure of tokenized inputs. This work presents a theoretical and empirical investigation into how tokenization schemes, particularly subword-based methods like byte-pair encoding (BPE), impede symbolic computation by merging or obscuring atomic reasoning units. We introduce the notio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14178","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/2505.14178/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":"2505.14178","created_at":"2026-07-05T11:05:57.249647+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14178v1","created_at":"2026-07-05T11:05:57.249647+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14178","created_at":"2026-07-05T11:05:57.249647+00:00"},{"alias_kind":"pith_short_12","alias_value":"A2SFIPNYOMYH","created_at":"2026-07-05T11:05:57.249647+00:00"},{"alias_kind":"pith_short_16","alias_value":"A2SFIPNYOMYHJVAD","created_at":"2026-07-05T11:05:57.249647+00:00"},{"alias_kind":"pith_short_8","alias_value":"A2SFIPNY","created_at":"2026-07-05T11:05:57.249647+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/A2SFIPNYOMYHJVADXEGLFDTADL","json":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL.json","graph_json":"https://pith.science/api/pith-number/A2SFIPNYOMYHJVADXEGLFDTADL/graph.json","events_json":"https://pith.science/api/pith-number/A2SFIPNYOMYHJVADXEGLFDTADL/events.json","paper":"https://pith.science/paper/A2SFIPNY"},"agent_actions":{"view_html":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL","download_json":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL.json","view_paper":"https://pith.science/paper/A2SFIPNY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14178&json=true","fetch_graph":"https://pith.science/api/pith-number/A2SFIPNYOMYHJVADXEGLFDTADL/graph.json","fetch_events":"https://pith.science/api/pith-number/A2SFIPNYOMYHJVADXEGLFDTADL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL/action/storage_attestation","attest_author":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL/action/author_attestation","sign_citation":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL/action/citation_signature","submit_replication":"https://pith.science/pith/A2SFIPNYOMYHJVADXEGLFDTADL/action/replication_record"}},"created_at":"2026-07-05T11:05:57.249647+00:00","updated_at":"2026-07-05T11:05:57.249647+00:00"}