{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T04:07:15Z","timestamp":1779422835412,"version":"3.53.1"},"publisher-location":"New York, NY, USA","reference-count":10,"publisher":"ACM","license":[{"start":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T00:00:00Z","timestamp":1779753600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,5,26]]},"DOI":"10.1145\/3786335.3813194","type":"proceedings-article","created":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T03:16:22Z","timestamp":1779419782000},"page":"1327-1331","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Behavioral Fingerprints for LLM Endpoint Stability and Identity"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6892-8724","authenticated-orcid":false,"given":"Jonah","family":"Leshin","sequence":"first","affiliation":[{"name":"VAIL, San Francisco, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-1339-3153","authenticated-orcid":false,"given":"Manish","family":"Shah","sequence":"additional","affiliation":[{"name":"VAIL, San Francisco, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5116-1421","authenticated-orcid":false,"given":"Ian","family":"Timmis","sequence":"additional","affiliation":[{"name":"VAIL, San Francisco, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9860-9938","authenticated-orcid":false,"given":"Daniel","family":"Kang","sequence":"additional","affiliation":[{"name":"University of Illinois at Urbana-Champaign, Urbana, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,5,26]]},"reference":[{"key":"e_1_3_3_3_2_2","unstructured":"Florian Brand and Jean-Stanislas Denain. 2025. Why Benchmarking is Hard. https:\/\/epoch.ai\/gradient-updates\/why-benchmarking-is-hard\/. Accessed: 2026-03-10."},{"key":"e_1_3_3_3_3_2","unstructured":"Timoth\u00e9e Chauvin Cl\u00e9ment Lalanne Erwan Le\u00a0Merrer Jean-Michel Loubes Fran\u00e7ois Ta\u00efani and Gilles Tredan. 2026. Token-Efficient Change Detection in LLM APIs. arXiv preprint arXiv:https:\/\/arXiv.org\/abs\/2602.11083 (2026)."},{"key":"e_1_3_3_3_4_2","doi-asserted-by":"publisher","unstructured":"Lingjiao Chen Matei Zaharia and James Zou. 2024. How Is ChatGPT\u2019s Behavior Changing over Time? Harvard Data Science Review 6 2 (2024). 10.1162\/99608f92.5317da47https:\/\/hdsr.mitpress.mit.edu\/pub\/y95zitmz.","DOI":"10.1162\/99608f92.5317da47"},{"key":"e_1_3_3_3_5_2","doi-asserted-by":"publisher","unstructured":"Horace He and Thinking Machines Lab. 2025. Defeating Nondeterminism in LLM Inference. Thinking Machines Lab: Connectionism (2025). 10.64434\/tml.20250910https:\/\/thinkingmachines.ai\/blog\/defeating-nondeterminism-in-llm-inference\/.","DOI":"10.64434\/tml.20250910"},{"key":"e_1_3_3_3_6_2","unstructured":"Margin Research. 2025. Claude Code Historical Performance Tracker. https:\/\/marginlab.ai\/trackers\/claude-code-historical-performance\/."},{"key":"e_1_3_3_3_7_2","unstructured":"Jacob Phillips. 2025. daily-bench. https:\/\/github.com\/jacobphillips99\/daily-bench."},{"key":"e_1_3_3_3_8_2","doi-asserted-by":"publisher","DOI":"10.1561\/3600000002"},{"key":"e_1_3_3_3_9_2","doi-asserted-by":"publisher","unstructured":"G\u00e1bor\u00a0J. Sz\u00e9kely and Maria\u00a0L. Rizzo. 2013. Energy Statistics: A Class of Statistics Based on Distances. Journal of Statistical Planning and Inference 143 8 (2013) 1249\u20131272. 10.1016\/j.jspi.2013.03.018","DOI":"10.1016\/j.jspi.2013.03.018"},{"key":"e_1_3_3_3_10_2","doi-asserted-by":"crossref","unstructured":"Jingze Wang and Nan Wang. 2025. Assessing Consistency and Reproducibility in the Outputs of Large Language Models: Evidence Across Diverse Finance and Accounting Tasks. arXiv preprint arXiv:https:\/\/arXiv.org\/abs\/2503.16974 (2025).","DOI":"10.2139\/ssrn.5189069"},{"key":"e_1_3_3_3_11_2","volume-title":"A Tiny Review on E-values and E-processes","author":"Wang Ruodu","year":"2023","unstructured":"Ruodu Wang. 2023. A Tiny Review on E-values and E-processes. Technical Report. University of Waterloo."}],"event":{"name":"CAIS '26: ACM Conference on AI and Agentic Systems","location":"San Jose CA USA","acronym":"CAIS '26"},"container-title":["Proceedings of the ACM Conference on AI and Agentic Systems"],"original-title":[],"deposited":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T03:23:02Z","timestamp":1779420182000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3786335.3813194"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,26]]},"references-count":10,"alternative-id":["10.1145\/3786335.3813194","10.1145\/3786335"],"URL":"https:\/\/doi.org\/10.1145\/3786335.3813194","relation":{},"subject":[],"published":{"date-parts":[[2026,5,26]]},"assertion":[{"value":"2026-05-26","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}