{"id":"tp-37631","type":"article-newspaper","title":"Perplexity's JOLT Study Estimates 6.5-Fold Lower Training Token Use","author":"Simon Yoon","publisher":"TOKENPOST Inc.","url":"https://www.tokenpost.com/news/technology/37631","datePublished":"2026-10-22T08:05:12.000Z","dateModified":"2026-10-22T08:05:12.000Z","language":"en","abstract":"At 63.44% accuracy on MATH, JOLT's fitted budget was 1.30 million processed tokens, compared with 8.40 million at an observed GRPO checkpoint.","section":"Technology","references":[]}