"How much do large language models memorize" by @jxmnop et al. is a fantastic paper that received one of the best paper awards at ICML 2026 I took …

Title: How much do large language models memorize Source: rss_twitter An ICML 2026 paper quantifies memorization levels in large language models through empirical testing. Reproducing the core claims at a reduced scale requires less than $8 in GPU compute, providing a cost-effective method for reviewers and practitioners to validate memorization metrics.