{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YVLLQUFLYQ5IARVFRFUNYX4EQH","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"22dee1d81089a51a767edbb4cc74cbe1aff0ce3c780e9934232ed5424f06c492","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-01T17:55:47Z","title_canon_sha256":"304992a20482eb25a54260df7b77e4d59301518ef19be62c90933ffa0fdc90bd"},"schema_version":"1.0","source":{"id":"2605.00817","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.00817","created_at":"2026-05-22T01:04:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.00817v2","created_at":"2026-05-22T01:04:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.00817","created_at":"2026-05-22T01:04:03Z"},{"alias_kind":"pith_short_12","alias_value":"YVLLQUFLYQ5I","created_at":"2026-05-22T01:04:03Z"},{"alias_kind":"pith_short_16","alias_value":"YVLLQUFLYQ5IARVF","created_at":"2026-05-22T01:04:03Z"},{"alias_kind":"pith_short_8","alias_value":"YVLLQUFL","created_at":"2026-05-22T01:04:03Z"}],"graph_snapshots":[{"event_id":"sha256:5d9c3f3609b5b4a1dd558b4d5c61245007dc52d799273ad631744be0496a675b","target":"graph","created_at":"2026-05-22T01:04:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Across 14 models and 55 datasets, average first-answer accuracy drops from 61% on 5-step procedures to 20% on 95-step procedures."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps."}],"snapshot_sha256":"3530dc561fd32349a47649a7afb64426ca96dddf22040cb3ec74195d0bcd8f41"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2605.00817/integrity.json","findings":[],"snapshot_sha256":"e2155f906a7b2a39fc3bb17418856f5e943248c4e7b4e1c5ceb753f3797447b2","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Abhishek Upperwal, Mayank Singh, Pritam Kadasi, Sailesh Panda","cross_cats":[],"headline":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-01T17:55:47Z","title":"When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.00817","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-09T18:48:26.279158Z","id":"d170fc2a-8b53-4efc-a5b0-532cb5c232c9","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"LLMs lose accuracy on following multi-step arithmetic procedures as length grows from 5 to 95 steps.","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.","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."}},"verdict_id":"d170fc2a-8b53-4efc-a5b0-532cb5c232c9"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8107d9b71281bc227ad02d477f406c78bb8c6093eaa5aea3472ba2838e8435c5","target":"record","created_at":"2026-05-22T01:04:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"22dee1d81089a51a767edbb4cc74cbe1aff0ce3c780e9934232ed5424f06c492","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-01T17:55:47Z","title_canon_sha256":"304992a20482eb25a54260df7b77e4d59301518ef19be62c90933ffa0fdc90bd"},"schema_version":"1.0","source":{"id":"2605.00817","kind":"arxiv","version":2}},"canonical_sha256":"c556b850abc43a8046a58968dc5f8481c1fa3e0421bd2948b31121bbb0fa1dd5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c556b850abc43a8046a58968dc5f8481c1fa3e0421bd2948b31121bbb0fa1dd5","first_computed_at":"2026-05-22T01:04:03.936329Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:03.936329Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wpJXG83h9OwHgQTp93GFwynd8gCxeJTvqc7pIRndj5N6RQ9b6DBaAL097bnoNLNbhQkhvqNABzi+jUhrZckgCQ==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:03.937304Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.00817","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8107d9b71281bc227ad02d477f406c78bb8c6093eaa5aea3472ba2838e8435c5","sha256:5d9c3f3609b5b4a1dd558b4d5c61245007dc52d799273ad631744be0496a675b"],"state_sha256":"5495abe02e345e36131de1b86bc2cdc911d720e78cbe5c26f5f1df3a868445be"}