{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YVLLQUFLYQ5IARVFRFUNYX4EQH","short_pith_number":"pith:YVLLQUFL","schema_version":"1.0","canonical_sha256":"c556b850abc43a8046a58968dc5f8481c1fa3e0421bd2948b31121bbb0fa1dd5","source":{"kind":"arxiv","id":"2605.00817","version":2},"attestation_state":"computed","paper":{"title":"When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhishek Upperwal, Mayank Singh, Pritam Kadasi, Sailesh Panda","submitted_at":"2026-05-01T17:55:47Z","abstract_excerpt":"Large language models (LLMs) often achieve strong performance on reasoning benchmarks, but final-answer accuracy alone does not show whether they faithfully execute the procedure specified in a prompt. We study this question through a controlled diagnostic benchmark for procedural execution, where models are given a step-wise arithmetic algorithm and two numeric inputs, and must return the final computed value. The benchmark uses simple arithmetic operations but increases complexity through algorithm length and look-back dependencies over intermediate variables. Across 14 models and 55 dataset"},"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":"2605.00817","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-01T17:55:47Z","cross_cats_sorted":[],"title_canon_sha256":"304992a20482eb25a54260df7b77e4d59301518ef19be62c90933ffa0fdc90bd","abstract_canon_sha256":"22dee1d81089a51a767edbb4cc74cbe1aff0ce3c780e9934232ed5424f06c492"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:03.937304Z","signature_b64":"wpJXG83h9OwHgQTp93GFwynd8gCxeJTvqc7pIRndj5N6RQ9b6DBaAL097bnoNLNbhQkhvqNABzi+jUhrZckgCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c556b850abc43a8046a58968dc5f8481c1fa3e0421bd2948b31121bbb0fa1dd5","last_reissued_at":"2026-05-22T01:04:03.936329Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:03.936329Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhishek Upperwal, Mayank Singh, Pritam Kadasi, Sailesh Panda","submitted_at":"2026-05-01T17:55:47Z","abstract_excerpt":"Large language models (LLMs) often achieve strong performance on reasoning benchmarks, but final-answer accuracy alone does not show whether they faithfully execute the procedure specified in a prompt. We study this question through a controlled diagnostic benchmark for procedural execution, where models are given a step-wise arithmetic algorithm and two numeric inputs, and must return the final computed value. The benchmark uses simple arithmetic operations but increases complexity through algorithm length and look-back dependencies over intermediate variables. Across 14 models and 55 dataset"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across 14 models and 55 datasets, average first-answer accuracy drops from 61% on 5-step procedures to 20% on 95-step procedures.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the measured accuracy drop reflects failures of procedural fidelity rather than unrelated factors such as context-window limits, arithmetic capability, or prompt formatting.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LLM accuracy on controlled procedural arithmetic drops from 61% at 5 steps to 20% at 95 steps, with failures including skipped steps, premature answers, and hallucinated operations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3530dc561fd32349a47649a7afb64426ca96dddf22040cb3ec74195d0bcd8f41"},"source":{"id":"2605.00817","kind":"arxiv","version":2},"verdict":{"id":"d170fc2a-8b53-4efc-a5b0-532cb5c232c9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T18:48:26.279158Z","strongest_claim":"Across 14 models and 55 datasets, average first-answer accuracy drops from 61% on 5-step procedures to 20% on 95-step procedures.","one_line_summary":"LLM accuracy on controlled procedural arithmetic drops from 61% at 5 steps to 20% at 95 steps, with failures including skipped steps, premature answers, and hallucinated operations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the measured accuracy drop reflects failures of procedural fidelity rather than unrelated factors such as context-window limits, arithmetic capability, or prompt formatting.","pith_extraction_headline":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.00817/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T18:41:01.458485Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:48:49.841220Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"e2155f906a7b2a39fc3bb17418856f5e943248c4e7b4e1c5ceb753f3797447b2"},"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":"2605.00817","created_at":"2026-05-22T01:04:03.936439+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.00817v2","created_at":"2026-05-22T01:04:03.936439+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.00817","created_at":"2026-05-22T01:04:03.936439+00:00"},{"alias_kind":"pith_short_12","alias_value":"YVLLQUFLYQ5I","created_at":"2026-05-22T01:04:03.936439+00:00"},{"alias_kind":"pith_short_16","alias_value":"YVLLQUFLYQ5IARVF","created_at":"2026-05-22T01:04:03.936439+00:00"},{"alias_kind":"pith_short_8","alias_value":"YVLLQUFL","created_at":"2026-05-22T01:04:03.936439+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/YVLLQUFLYQ5IARVFRFUNYX4EQH","json":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH.json","graph_json":"https://pith.science/api/pith-number/YVLLQUFLYQ5IARVFRFUNYX4EQH/graph.json","events_json":"https://pith.science/api/pith-number/YVLLQUFLYQ5IARVFRFUNYX4EQH/events.json","paper":"https://pith.science/paper/YVLLQUFL"},"agent_actions":{"view_html":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH","download_json":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH.json","view_paper":"https://pith.science/paper/YVLLQUFL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.00817&json=true","fetch_graph":"https://pith.science/api/pith-number/YVLLQUFLYQ5IARVFRFUNYX4EQH/graph.json","fetch_events":"https://pith.science/api/pith-number/YVLLQUFLYQ5IARVFRFUNYX4EQH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH/action/storage_attestation","attest_author":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH/action/author_attestation","sign_citation":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH/action/citation_signature","submit_replication":"https://pith.science/pith/YVLLQUFLYQ5IARVFRFUNYX4EQH/action/replication_record"}},"created_at":"2026-05-22T01:04:03.936439+00:00","updated_at":"2026-05-22T01:04:03.936439+00:00"}