{"as_of":"2026-08-16T02:39:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:8c7b1f97901613c3fd4a451273ce6b9dd0dc98c3451978ce0c674c7dea534db6","coverage":[{"denominator":23,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":23,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T16:17:47.859203Z","state":"measured"},{"denominator":23,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":23,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2608.13129/citation-record","integrity":"/paper/2608.13129/integrity","json":"/paper/2608.13129/citation-record.json","paper":"/paper/2608.13129"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2402.00157","last_updated":"2024-09-16T19:20:59Z","snapshot_observed_at":"2026-08-14T23:03:16.281379Z","submitted_at":"2024-01-31T20:26:32Z","title":"Large Language Models for Mathematical Reasoning: Progresses and Challenges","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.00157","snapshot_observed_at":"2026-08-15T16:17:47.556851Z","title":"Large language models for mathematical reasoning: Progresses and challenges.arXiv preprint arXiv:2402.00157,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.556851Z"},"links":{"cited_paper":"/paper/2402.00157","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:e5bc8de51bb1ec4cfcc02e127f86d88732a0766542da8c5fd2f7c938f1c38d34","observation_id":"7b06d7fb-b5f8-4bdc-8053-5107dd05a0e1","resolution":{"observed_at":"2026-08-15T16:17:47.556851Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.14168","last_updated":"2021-11-18T00:23:45Z","snapshot_observed_at":"2026-08-14T02:43:01.480086Z","submitted_at":"2021-10-27T04:49:45Z","title":"Training Verifiers to Solve Math Word Problems","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.14168","snapshot_observed_at":"2026-08-15T16:17:47.605797Z","title":"Training verifiers to solve math word problems.arXiv preprint arXiv:2110.14168,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.605797Z"},"links":{"cited_paper":"/paper/2110.14168","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:0fdc13a56099545f2e463f8c1b12d318bffaa89127fd16aeac55b80afc727386","observation_id":"10660d69-121f-4372-93e8-4f3bf446ab57","resolution":{"observed_at":"2026-08-15T16:17:47.605797Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.02207","last_updated":"2024-03-04T18:25:29Z","snapshot_observed_at":"2026-08-14T10:35:42.683947Z","submitted_at":"2023-10-03T17:06:52Z","title":"Language Models Represent Space and Time","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.02207","snapshot_observed_at":"2026-08-15T16:17:47.654981Z","title":"Languagemodelsrepresentspaceandtime.arXiv preprint arXiv:2310.02207,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.654981Z"},"links":{"cited_paper":"/paper/2310.02207","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:3d38c3dd03537568abd28884cece276076cf68fc31401c64f08ab0c03c71c5a6","observation_id":"653e56fd-e6b2-464a-97b0-c6e13a14f11a","resolution":{"observed_at":"2026-08-15T16:17:47.654981Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.01728","last_updated":"2024-01-29T06:27:53Z","snapshot_observed_at":"2026-08-13T07:00:35.291225Z","submitted_at":"2023-10-03T01:31:25Z","title":"Time-LLM: Time Series Forecasting by Reprogramming Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.01728","snapshot_observed_at":"2026-08-15T16:17:47.678936Z","title":"Time-llm: Timeseriesforecastingbyreprogramminglargelanguage models.arXiv preprint arXiv:2310.01728,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.678936Z"},"links":{"cited_paper":"/paper/2310.01728","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:269b037f859b96ca582ca1f6f6bab87ab82d662d6d4bff8cabf03be49d2928e8","observation_id":"ca6c5c39-b24d-4ce7-94d6-d431765c593b","resolution":{"observed_at":"2026-08-15T16:17:47.678936Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.20050","last_updated":"2023-05-31T17:24:00Z","snapshot_observed_at":"2026-08-11T17:22:43.545531Z","submitted_at":"2023-05-31T17:24:00Z","title":"Let's Verify Step by Step","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.20050","snapshot_observed_at":"2026-08-15T16:17:47.687123Z","title":"Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.687123Z"},"links":{"cited_paper":"/paper/2305.20050","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:fd148c4321c53b6aab41f6767366dca1676ce0ae08e41ef1cc4c842bdbcab984","observation_id":"be439f0b-5a90-4fb6-b8a0-38df3a2fbd67","resolution":{"observed_at":"2026-08-15T16:17:47.687123Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2307.10485","last_updated":"2023-11-14T16:34:00Z","snapshot_observed_at":"2026-08-13T10:52:16.351105Z","submitted_at":"2023-07-19T22:43:57Z","title":"FinGPT: Democratizing Internet-scale Data for Financial Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.10485","snapshot_observed_at":"2026-08-15T16:17:47.694066Z","title":"Fingpt: Democratizing internet-scale data for financial large language models.arXiv preprint arXiv:2307.10485,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.694066Z"},"links":{"cited_paper":"/paper/2307.10485","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:95b70df56af65596978f5a3a2f2ad4ef3e1742eee6a40411a4390d8861b5b799","observation_id":"32b4bf20-268b-4858-8c61-0e73ce9e1d8d","resolution":{"observed_at":"2026-08-15T16:17:47.694066Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.05229","last_updated":"2025-08-27T16:24:39Z","snapshot_observed_at":"2026-08-15T01:09:53.276153Z","submitted_at":"2024-10-07T17:36:37Z","title":"GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.05229","snapshot_observed_at":"2026-08-15T16:17:47.719372Z","title":"GSM-Symbolic: Understanding the limitations of mathematical reasoning in large language models.arXiv preprint arXiv:2410.05229,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.719372Z"},"links":{"cited_paper":"/paper/2410.05229","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:815c50c3bc3c83a661dd41e04cd8e69c37075d3e35a0e6a3da836a440e205acb","observation_id":"9edbd3f0-7f23-44f6-9c88-02bf4a3250be","resolution":{"observed_at":"2026-08-15T16:17:47.719372Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2102.13019","last_updated":"2021-04-12T19:58:27Z","snapshot_observed_at":"2026-08-12T04:51:09.122205Z","submitted_at":"2021-02-25T17:22:53Z","title":"Investigating the Limitations of Transformers with Simple Arithmetic Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2102.13019","snapshot_observed_at":"2026-08-15T16:17:47.732766Z","title":"Investigating the limitations of transformers with simple arithmetic tasks.arXiv preprint arXiv:2102.13019,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.732766Z"},"links":{"cited_paper":"/paper/2102.13019","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:51cd4efc5af7722219a9ded0a27ee904d2fedd596f1346004003cf1ff51ca512","observation_id":"c2972dbe-339a-4972-b7a5-f4677ee152aa","resolution":{"observed_at":"2026-08-15T16:17:47.732766Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2112.00114","last_updated":"2021-11-30T21:32:46Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2021-11-30T21:32:46Z","title":"Show Your Work: Scratchpads for Intermediate Computation with Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2112.00114","snapshot_observed_at":"2026-08-15T16:17:47.748806Z","title":"Show your work: Scratchpads for intermediate computation with language models.arXiv preprint arXiv:2112.00114,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.748806Z"},"links":{"cited_paper":"/paper/2112.00114","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:1720323858100ae8ef19889a21e85a2592867e1661badad2c56706d3dadc9e3d","observation_id":"4a4c436f-f42b-4fe5-af57-c670a212bcfc","resolution":{"observed_at":"2026-08-15T16:17:47.748806Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.00071","last_updated":"2026-02-06T19:40:50Z","snapshot_observed_at":"2026-08-01T02:15:47.181936Z","submitted_at":"2023-08-31T18:18:07Z","title":"YaRN: Efficient Context Window Extension of Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.00071","snapshot_observed_at":"2026-08-15T16:17:47.795647Z","title":"Bowen Peng, Jeffrey Quesnelle, Honglu Fan, and Enrico Shippole","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.795647Z"},"links":{"cited_paper":"/paper/2309.00071","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:557e883558bbff5fc2d1e2052a77621f1a96b2f3b05e51872928bb3a0938fc09","observation_id":"2d9833d8-87f4-46cd-b425-2c7ccb9f1b07","resolution":{"observed_at":"2026-08-15T16:17:47.795647Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.03300","last_updated":"2024-04-27T15:25:53Z","snapshot_observed_at":"2026-08-06T14:58:42.911363Z","submitted_at":"2024-02-05T18:55:32Z","title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.03300","snapshot_observed_at":"2026-08-15T16:17:47.808664Z","title":"Deepseekmath: Pushing the limits of mathematical reasoning in open language models.arXiv preprint arXiv:2402.03300,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.808664Z"},"links":{"cited_paper":"/paper/2402.03300","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:a0e1286fa87d2e69b995392e82bd0d565fd8c18794fe4fbebc7c5ff284adab79","observation_id":"44add5cc-345b-4452-a27b-5c0a2b5dad3f","resolution":{"observed_at":"2026-08-15T16:17:47.808664Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.02884","last_updated":"2024-03-05T11:42:59Z","snapshot_observed_at":"2026-08-14T08:08:59.449816Z","submitted_at":"2024-03-05T11:42:59Z","title":"MathScale: Scaling Instruction Tuning for Mathematical Reasoning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.02884","snapshot_observed_at":"2026-08-15T16:17:47.815337Z","title":"MathScale: Scaling instruction tuning for mathematical reasoning.arXiv preprint arXiv:2403.02884,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.815337Z"},"links":{"cited_paper":"/paper/2403.02884","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:8964f888549cb8aa18ecb3ee401e596d0bfdff9facd213cd09e6f37d18a1049d","observation_id":"c9f5ede2-2d16-493f-8e4d-5088cc97eab6","resolution":{"observed_at":"2026-08-15T16:17:47.815337Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.09085","last_updated":"2022-11-16T18:06:33Z","snapshot_observed_at":"2026-08-13T16:29:32.694746Z","submitted_at":"2022-11-16T18:06:33Z","title":"Galactica: A Large Language Model for Science","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2211.09085","snapshot_observed_at":"2026-08-15T16:17:47.826591Z","title":"Galactica: A large language model for science.arXiv preprint arXiv:2211.09085,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.826591Z"},"links":{"cited_paper":"/paper/2211.09085","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:c8eb22c6ee56cb0af11ec06844311da42d7e020fd79dfda675a86e46fc909ac3","observation_id":"1ffe5a15-6abf-401c-b171-0b77ec014f43","resolution":{"observed_at":"2026-08-15T16:17:47.826591Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.14275","last_updated":"2022-11-25T18:19:44Z","snapshot_observed_at":"2026-08-01T02:16:43.109337Z","submitted_at":"2022-11-25T18:19:44Z","title":"Solving math word problems with process- and outcome-based feedback","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2211.14275","snapshot_observed_at":"2026-08-15T16:17:47.839665Z","title":"Solvingmathwordproblemswithprocess-andoutcome-basedfeedback","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.839665Z"},"links":{"cited_paper":"/paper/2211.14275","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:4be510bb075c7fa9c03a58c9d22357b193088b9860407c6dcb87d8af6b232f62","observation_id":"209ea593-85d5-4ac5-b81c-daf6192d45ef","resolution":{"observed_at":"2026-08-15T16:17:47.839665Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.03766","last_updated":"2025-03-05T09:52:30Z","snapshot_observed_at":"2026-08-12T22:06:39.206325Z","submitted_at":"2024-11-06T08:59:44Z","title":"Number Cookbook: Number Understanding of Language Models and How to Improve It","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.03766","snapshot_observed_at":"2026-08-15T16:17:47.846277Z","title":"Number cookbook: Number under- standing of language models and how to improve it.arXiv preprint arXiv:2411.03766,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.846277Z"},"links":{"cited_paper":"/paper/2411.03766","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:793278bf0b6542e2acc9489ba98ffe6c7f3eff6d6131c79331a2c7b3c5805b3a","observation_id":"0225a707-0559-4b98-8a24-61ccf1d970c6","resolution":{"observed_at":"2026-08-15T16:17:47.846277Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.17564","last_updated":"2023-12-21T06:21:11Z","snapshot_observed_at":"2026-08-14T12:54:48.492396Z","submitted_at":"2023-03-30T17:30:36Z","title":"BloombergGPT: A Large Language Model for Finance","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.17564","snapshot_observed_at":"2026-08-15T16:17:47.859203Z","title":"Emergent abilities of large language models.Trans- actions on Machine Learning Research, 2022a","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.859203Z"},"links":{"cited_paper":"/paper/2303.17564","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:5123499e0b35a27e33ad3bb32a0acf24c03ea610120875bab6106b55a7214d14","observation_id":"2f40f32f-55b3-4513-ac66-2457a082f870","resolution":{"observed_at":"2026-08-15T16:17:47.859203Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2103.03874","last_updated":"2021-11-08T21:30:18Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2021-03-05T18:59:39Z","title":"Measuring Mathematical Problem Solving With the MATH Dataset","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2103.03874","snapshot_observed_at":"2026-08-15T16:17:47.665861Z","title":"Measuring mathematical problem solving with the MATH dataset.arXiv preprint arXiv:2103.03874,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":1990,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.665861Z"},"links":{"cited_paper":"/paper/2103.03874","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:a23b100d6fb5e62f454cb3ba779a068e99c7c894d6ac478a328a14fc1bf2cc2c","observation_id":"70939265-7a32-4bc5-bce6-d759deca8545","resolution":{"observed_at":"2026-08-15T16:17:47.665861Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.21783","last_updated":"2024-11-23T23:27:33Z","snapshot_observed_at":"2026-08-13T17:20:44.002518Z","submitted_at":"2024-07-31T17:54:27Z","title":"The Llama 3 Herd of Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.21783","snapshot_observed_at":"2026-08-15T16:17:47.630256Z","title":"The Llama 3 herd of models.arXiv preprint arXiv:2407.21783,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2011,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.630256Z"},"links":{"cited_paper":"/paper/2407.21783","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:929a0e986c466485c8146967918bc5d26e2f15ec0f83cdad1d3e8e7d5e3c5e6d","observation_id":"ad028776-5aac-4b20-8f64-281fd83c127f","resolution":{"observed_at":"2026-08-15T16:17:47.630256Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.12948","last_updated":"2026-01-04T03:57:36Z","snapshot_observed_at":"2026-08-15T12:33:55.451951Z","submitted_at":"2025-01-22T15:19:35Z","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.12948","snapshot_observed_at":"2026-08-15T16:17:47.613609Z","title":"DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning.arXiv preprint arXiv:2501.12948,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.613609Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:06e31698c030a30a3b15c4892449afb3db710311a1f9ce8e279eba3b375a0245","observation_id":"83ca18ed-267d-4d0c-ba39-c6da0220b4e5","resolution":{"observed_at":"2026-08-15T16:17:47.613609Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.05332","last_updated":"2023-04-11T16:50:17Z","snapshot_observed_at":"2026-07-06T15:14:28.574778Z","submitted_at":"2023-04-11T16:50:17Z","title":"Emergent autonomous scientific research capabilities of large language models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.05332","snapshot_observed_at":"2026-08-15T16:17:47.571088Z","title":"Emergent autonomous scientific research capabilities of large language models.arXiv preprint arXiv:2304.05332,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.571088Z"},"links":{"cited_paper":"/paper/2304.05332","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:1f21b09f00b85d2cd5eb97369ce75e8804543e21a247e0390f7174b847ab80c8","observation_id":"fabd9102-37f0-4681-84e0-ca9ee6280816","resolution":{"observed_at":"2026-08-15T16:17:47.571088Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:17:49.637907Z","title":"xVal: A continuous number encoding for large language models","venue":null,"work_id":"c9aae660-4e69-4a0e-a056-23805131f58c","year":2023},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.639463Z"},"links":{"citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:431fc700d0363dd14114b326935b1792b78c40da0d5ecb5a18387d844b94378e","observation_id":"9363f353-05d8-45e4-8d6c-6b8e688ad147","resolution":{"observed_at":"2026-08-15T16:17:49.654075Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.17399","last_updated":"2024-12-23T12:46:06Z","snapshot_observed_at":"2026-08-12T23:56:52.518306Z","submitted_at":"2024-05-27T17:49:18Z","title":"Transformers Can Do Arithmetic with the Right Embeddings","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.17399","snapshot_observed_at":"2026-08-15T16:17:47.704904Z","title":"Transformers can do arithmetic with the right embeddings.arXiv preprint arXiv:2405.17399,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.704904Z"},"links":{"cited_paper":"/paper/2405.17399","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:dabff0beba6e8b43ad48412a6a7514ccbe3324216fedc7de47d491849cd65010","observation_id":"1ac7a6e7-5c3e-4039-970a-fd576ab5c714","resolution":{"observed_at":"2026-08-15T16:17:47.704904Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.05624","last_updated":"2022-06-07T07:37:30Z","snapshot_observed_at":"2026-08-13T16:59:03.539050Z","submitted_at":"2022-01-14T19:05:44Z","title":"Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.05624","snapshot_observed_at":"2026-08-15T16:17:47.579889Z","title":"Learning the greatest common divisor: explaining transformer predictions.arXiv preprint arXiv:2201.05624,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement","version":1},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-15T16:17:47.579889Z"},"links":{"cited_paper":"/paper/2201.05624","citing_paper":"/paper/2608.13129"},"observation_digest":"sha256:8bd844ec5c68f99b205cc146e60939fb43f54dc43c718b39ae39a1876c831c5e","observation_id":"9287a38c-ce4a-456e-aa35-3325e8ccd5d4","resolution":{"observed_at":"2026-08-15T16:17:47.579889Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2608.13129","last_updated":"2026-08-13T12:01:58Z","latest_version":1,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-16T02:11:32.446489Z","submitted_at":"2026-08-13T12:01:58Z","title":"Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement"},"reference_resolution":{"displayed":23,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":22,"verified_exact":0,"verified_fuzzy":1},"total_outbound_references":23},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 16 August 2026, this Paper Citation Record lists 23 of 23 outbound references and 0 inbound Pith citation observations for arXiv:2608.13129."}