{"id":"ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf","title":"What multiplication errors reveal about LLM arithmetic","argument":"William Bolton, Bezawit Tilaye and Huw Day report that multiplication errors in the small base LLMs they tested do not follow the “noisy quantisation” pattern previously described for addition. They examined 16,991 problems with varied operand lengths. In small Qwen models, they found that lower-order columns—digits representing smaller place values—predominantly preserved parity, meaning whether a number is even or odd. In higher-order digits, errors shifted toward off-by-one carry failures: mistakes involving a carry to the next place value that is one away from the expected result.\n\nThe authors suggest these patterns reflect distinct, multi-periodic internal representations, with parity tracked more robustly than exact digit identity. That is a proposed interpretation of the error patterns, not evidence that the models perform multiplication reliably. The findings offer a way to investigate how numerical reasoning in transformers may differ across operations and digit positions. The abstract does not establish whether the same patterns hold in larger or other model families, or how the errors affect real-world uses. Do these multiplication error patterns persist beyond the small Qwen models examined?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The abstract describes tests across several small base LLMs and identifies the specific error regime in small Qwen models; it does not establish whether the patterns generalize to larger models or other model families. The abstract does not report how the observed errors affect real-world uses of LLMs.","speaker":{"name":"Bolton, William, Tilaye, Bezawit, Day, Huw","roleAtSource":"Article author"},"source":{"title":"\"Close enough?\" An investigation into multiplication errors in LLMs","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.38634","publishedAt":"2026-10-28","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors report that multiplication errors do not conform to the addition-style noisy quantisation model.","excerpt":"We show that multiplication errors do not conform to the addition-style noisy quantisation model.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"In small Qwen models, lower-order column errors predominantly preserve parity, while higher-order digit errors shift toward off-by-one carry failures.","excerpt":"errors in lower-order columns predominantly preserve parity. Conversely, higher-order digit errors shift toward off-by-one carry failures.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"The authors suggest that these patterns indicate distinct internal representations in which parity is tracked more robustly than exact digit identity.","excerpt":"arithmetic computation in LLMs relies on distinct, multi-periodic internal representations where parity is tracked more robustly than exact digit identity","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"The study evaluates multiplication across several small base LLMs using a dataset of 16,991 problems with varied operand lengths.","excerpt":"we evaluate multiplication capabilities across several small base LLMs on a dataset of 16,991 problems spanning varied operand lengths.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.38634","checkedAt":"2026-11-11T12:56:07.956Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do these multiplication error patterns persist beyond the small Qwen models examined?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"What multiplication errors reveal about LLM arithmetic","summary":"William Bolton, Bezawit Tilaye and Huw Day report that multiplication errors in the small base LLMs they tested do not follow the “noisy quantisation” pattern previously described for addition. They examined 16,991 problems with varied operand lengths. In small Qwen models, they found that lower-order columns—digits representing smaller place values—predominantly preserved parity, meaning whether a number is even or odd. In higher-order digits, errors shifted toward off-by-one carry failures: mistakes involving a carry to the next place value that is one away from the expected result.\n\nThe authors suggest these patterns reflect distinct, multi-periodic internal representations, with parity tracked more robustly than exact digit identity. That is a proposed interpretation of the error patterns, not evidence that the models perform multiplication reliably. The findings offer a way to investigate how numerical reasoning in transformers may differ across operations and digit positions. The abstract does not establish whether the same patterns hold in larger or other model families, or how the errors affect real-world uses. Do these multiplication error patterns persist beyond the small Qwen models examined?","question":"Do these multiplication error patterns persist beyond the small Qwen models examined?","claims":[{"text":"The authors report that multiplication errors do not conform to the addition-style noisy quantisation model.","excerpt":"We show that multiplication errors do not conform to the addition-style noisy quantisation model.","support":"source-stated"},{"text":"In small Qwen models, lower-order column errors predominantly preserve parity, while higher-order digit errors shift toward off-by-one carry failures.","excerpt":"errors in lower-order columns predominantly preserve parity. Conversely, higher-order digit errors shift toward off-by-one carry failures.","support":"source-stated"},{"text":"The authors suggest that these patterns indicate distinct internal representations in which parity is tracked more robustly than exact digit identity.","excerpt":"arithmetic computation in LLMs relies on distinct, multi-periodic internal representations where parity is tracked more robustly than exact digit identity","support":"source-stated"},{"text":"The study evaluates multiplication across several small base LLMs using a dataset of 16,991 problems with varied operand lengths.","excerpt":"we evaluate multiplication capabilities across several small base LLMs on a dataset of 16,991 problems spanning varied operand lengths.","support":"source-stated"}],"limitations":["The abstract describes tests across several small base LLMs and identifies the specific error regime in small Qwen models; it does not establish whether the patterns generalize to larger models or other model families.","The abstract does not report how the observed errors affect real-world uses of LLMs."],"mediaId":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","correctionReason":null,"id":"ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf","publicationId":"e497a73c-3651-40f2-8fab-954a304dd5b2","version":1,"sourceUrl":"https://arxiv.org/abs/2610.38634","sourceTitle":"\"Close enough?\" An investigation into multiplication errors in LLMs","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-28","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-30T06:45:33.466Z","createdAt":"2026-11-11T11:53:59.768Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","kind":"article","url":"https://arxiv.org/abs/2610.38634","sourceUrl":"https://arxiv.org/abs/2610.38634","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-31T10:49:44.009Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.38634","checkedAt":"2026-11-11T12:56:07.956Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","confidence":0.8300000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"794fab83445853d665fdd7f7e9b72d38939fa33c596ccd4ab5a2aab89920d2be","decidedAt":"2026-10-31T10:49:46.623Z"},"provenance":{"textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observationId":"e1d71e91ed0e71f4dc93429613b42d44ada94f55a049074d5475dbf26610e479","retrievedAt":"2026-10-31T10:49:43.787Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"f1500a31-c967-4d66-8c87-768a8a0c4290"},"reviewer":{"model":"gpt-6-luna","runId":"6c9f08d9-931b-4521-8008-c6f49884728d"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-11T11:53:59.768Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Bolton, William","url":null,"kind":"unknown"},{"name":"Tilaye, Bezawit","url":null,"kind":"unknown"},{"name":"Day, Huw","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Bolton, William","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Bolton, William","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}},{"actor":{"name":"Tilaye, Bezawit","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Tilaye, Bezawit","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}},{"actor":{"name":"Day, Huw","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Day, Huw","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}}],"participationCoverage":{"total":3,"returned":3,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.38634","snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z","basis":"publisher-page-metadata","sourceLinks":41,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf/source"}},"canonicalUrl":"https://escapemeridian.com/readings/what-multiplication-errors-reveal-about-llm-arithmetic-ca7f0246","formats":{"html":"/readings/what-multiplication-errors-reveal-about-llm-arithmetic-ca7f0246","markdown":"/readings/what-multiplication-errors-reveal-about-llm-arithmetic-ca7f0246.md","json":"/readings/what-multiplication-errors-reveal-about-llm-arithmetic-ca7f0246.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-llm","name":"Large language model","explanation":"A language model with many learned numerical settings, used to process and generate language.","referenceUrl":"https://developers.google.com/machine-learning/glossary#large-language-model"},{"id":"term-quantization","name":"Quantization","explanation":"Storing a model's numbers at lower precision so it uses less memory and runs faster, sometimes at a cost in accuracy.","referenceUrl":"https://developers.google.com/machine-learning/glossary#quantization"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf","revision":"e497a73c-3651-40f2-8fab-954a304dd5b2","revisionKind":"publication-id","title":"What multiplication errors reveal about LLM arithmetic","url":"https://escapemeridian.com/readings/what-multiplication-errors-reveal-about-llm-arithmetic-ca7f0246"},"source":{"type":"source","id":"https://arxiv.org/abs/2610.38634","revision":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","revisionKind":"sha256-source-text","title":"\"Close enough?\" An investigation into multiplication errors in LLMs","url":"https://arxiv.org/abs/2610.38634"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf","title":"What multiplication errors reveal about LLM arithmetic","argument":"William Bolton, Bezawit Tilaye and Huw Day report that multiplication errors in the small base LLMs they tested do not follow the “noisy quantisation” pattern previously described for addition. They examined 16,991 problems with varied operand lengths. In small Qwen models, they found that lower-order columns—digits representing smaller place values—predominantly preserved parity, meaning whether a number is even or odd. In higher-order digits, errors shifted toward off-by-one carry failures: mistakes involving a carry to the next place value that is one away from the expected result.\n\nThe authors suggest these patterns reflect distinct, multi-periodic internal representations, with parity tracked more robustly than exact digit identity. That is a proposed interpretation of the error patterns, not evidence that the models perform multiplication reliably. The findings offer a way to investigate how numerical reasoning in transformers may differ across operations and digit positions. The abstract does not establish whether the same patterns hold in larger or other model families, or how the errors affect real-world uses. Do these multiplication error patterns persist beyond the small Qwen models examined?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The abstract describes tests across several small base LLMs and identifies the specific error regime in small Qwen models; it does not establish whether the patterns generalize to larger models or other model families. The abstract does not report how the observed errors affect real-world uses of LLMs.","speaker":{"name":"Bolton, William, Tilaye, Bezawit, Day, Huw","roleAtSource":"Article author"},"source":{"title":"\"Close enough?\" An investigation into multiplication errors in LLMs","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.38634","publishedAt":"2026-10-28","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors report that multiplication errors do not conform to the addition-style noisy quantisation model.","excerpt":"We show that multiplication errors do not conform to the addition-style noisy quantisation model.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"In small Qwen models, lower-order column errors predominantly preserve parity, while higher-order digit errors shift toward off-by-one carry failures.","excerpt":"errors in lower-order columns predominantly preserve parity. Conversely, higher-order digit errors shift toward off-by-one carry failures.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"The authors suggest that these patterns indicate distinct internal representations in which parity is tracked more robustly than exact digit identity.","excerpt":"arithmetic computation in LLMs relies on distinct, multi-periodic internal representations where parity is tracked more robustly than exact digit identity","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"},{"text":"The study evaluates multiplication across several small base LLMs using a dataset of 16,991 problems with varied operand lengths.","excerpt":"we evaluate multiplication capabilities across several small base LLMs on a dataset of 16,991 problems spanning varied operand lengths.","url":"https://arxiv.org/abs/2610.38634","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.38634","checkedAt":"2026-11-11T12:56:07.956Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do these multiplication error patterns persist beyond the small Qwen models examined?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"What multiplication errors reveal about LLM arithmetic","summary":"William Bolton, Bezawit Tilaye and Huw Day report that multiplication errors in the small base LLMs they tested do not follow the “noisy quantisation” pattern previously described for addition. They examined 16,991 problems with varied operand lengths. In small Qwen models, they found that lower-order columns—digits representing smaller place values—predominantly preserved parity, meaning whether a number is even or odd. In higher-order digits, errors shifted toward off-by-one carry failures: mistakes involving a carry to the next place value that is one away from the expected result.\n\nThe authors suggest these patterns reflect distinct, multi-periodic internal representations, with parity tracked more robustly than exact digit identity. That is a proposed interpretation of the error patterns, not evidence that the models perform multiplication reliably. The findings offer a way to investigate how numerical reasoning in transformers may differ across operations and digit positions. The abstract does not establish whether the same patterns hold in larger or other model families, or how the errors affect real-world uses. Do these multiplication error patterns persist beyond the small Qwen models examined?","question":"Do these multiplication error patterns persist beyond the small Qwen models examined?","claims":[{"text":"The authors report that multiplication errors do not conform to the addition-style noisy quantisation model.","excerpt":"We show that multiplication errors do not conform to the addition-style noisy quantisation model.","support":"source-stated"},{"text":"In small Qwen models, lower-order column errors predominantly preserve parity, while higher-order digit errors shift toward off-by-one carry failures.","excerpt":"errors in lower-order columns predominantly preserve parity. Conversely, higher-order digit errors shift toward off-by-one carry failures.","support":"source-stated"},{"text":"The authors suggest that these patterns indicate distinct internal representations in which parity is tracked more robustly than exact digit identity.","excerpt":"arithmetic computation in LLMs relies on distinct, multi-periodic internal representations where parity is tracked more robustly than exact digit identity","support":"source-stated"},{"text":"The study evaluates multiplication across several small base LLMs using a dataset of 16,991 problems with varied operand lengths.","excerpt":"we evaluate multiplication capabilities across several small base LLMs on a dataset of 16,991 problems spanning varied operand lengths.","support":"source-stated"}],"limitations":["The abstract describes tests across several small base LLMs and identifies the specific error regime in small Qwen models; it does not establish whether the patterns generalize to larger models or other model families.","The abstract does not report how the observed errors affect real-world uses of LLMs."],"mediaId":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","correctionReason":null,"id":"ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf","publicationId":"e497a73c-3651-40f2-8fab-954a304dd5b2","version":1,"sourceUrl":"https://arxiv.org/abs/2610.38634","sourceTitle":"\"Close enough?\" An investigation into multiplication errors in LLMs","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-28","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-30T06:45:33.466Z","createdAt":"2026-11-11T11:53:59.768Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","kind":"article","url":"https://arxiv.org/abs/2610.38634","sourceUrl":"https://arxiv.org/abs/2610.38634","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-31T10:49:44.009Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.38634","checkedAt":"2026-11-11T12:56:07.956Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"ed9c2d831f103410d046b6e7b18cdb3ae4840febf36194e963d18b6d1323a2ae","confidence":0.8300000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"794fab83445853d665fdd7f7e9b72d38939fa33c596ccd4ab5a2aab89920d2be","decidedAt":"2026-10-31T10:49:46.623Z"},"provenance":{"textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observationId":"e1d71e91ed0e71f4dc93429613b42d44ada94f55a049074d5475dbf26610e479","retrievedAt":"2026-10-31T10:49:43.787Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"f1500a31-c967-4d66-8c87-768a8a0c4290"},"reviewer":{"model":"gpt-6-luna","runId":"6c9f08d9-931b-4521-8008-c6f49884728d"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-11T11:53:59.768Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Bolton, William","url":null,"kind":"unknown"},{"name":"Tilaye, Bezawit","url":null,"kind":"unknown"},{"name":"Day, Huw","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Bolton, William","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Bolton, William","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}},{"actor":{"name":"Tilaye, Bezawit","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Tilaye, Bezawit","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}},{"actor":{"name":"Day, Huw","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Day, Huw","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.38634","locator":"head/meta[name=author|citation_author]"},"assertionId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-11T12:56:07.956Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","sourceUrl":"https://arxiv.org/abs/2610.38634","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z"}}],"participationCoverage":{"total":3,"returned":3,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.38634","snapshotId":"a66b09228cfa7f27849ecd0e64772b2037859a213d07bf5e94d16fa01a95df0b","contentSha256":"a38c0c7cd97e5aaed6a449aa549706e862a0fa38317cfe78f0dad8be26d68863","textHash":"5da619edacbcc9fead6752788cd5c3a48f9531bcbb1e87b91b4829b93d1726bf","observedAt":"2026-11-11T12:56:07.956Z","basis":"publisher-page-metadata","sourceLinks":41,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/ca7f024694a43ef966e3f6ea2418edf49c855f977dde0c58ab3d06f34174ccbf/source"}}}],"audits":[{"publicationId":"e497a73c-3651-40f2-8fab-954a304dd5b2","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly names William Bolton, Bezawit Tilaye and Huw Day as the authors. It keeps the findings and the interpretation framed as theirs ('report', 'suggest').","excerpt":"Authors:William Bolton, Bezawit Tilaye, Huw Day"},{"dimension":"summary","status":"uncertain","explanation":"First, the central contrast term 'noisy quantisation' is never explained. A general reader cannot tell what multiplication fails to conform to. Smallest repair: add the source's own description, that addition errors are localised around carry boundaries. Second, the summary says 'lower-order columns... predominantly preserved parity'. The source says the errors in those columns preserve parity, so 'errors' should be restored. Third, the gloss on off-by-one carry failures is muddled, and 'multi-periodic' is left unexplained. Fourth, the closing question is repeated verbatim at the end of the summary.","excerpt":"addition errors follow a \"noisy quantisation\" model localised around carry boundaries"},{"dimension":"date","status":"pass","explanation":"The original published date, 2026-10-28, matches the v1 submission date on the arXiv page.","excerpt":"[Submitted on 28 Oct 2026]"},{"dimension":"support","status":"pass","explanation":"Each claim matches the abstract: the dataset size, the departure from the addition model, the parity and off-by-one pattern in small Qwen models, and the authors' suggested interpretation. The limitations on generalization and real-world impact are stated as unknowns rather than invented. The reading also keeps the source's suggestion separate from proof that the models multiply reliably.","excerpt":"Our findings suggest that arithmetic computation in LLMs relies on distinct, multi-periodic internal representations"},{"dimension":"media","status":"pass","explanation":"The selected media is the verified arXiv abstract page, which is the direct original source.","excerpt":null}],"createdAt":"2026-11-11T11:53:59.768Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}