<html aria-label="message body"><head><meta http-equiv="content-type" content="text/html; charset=us-ascii"></head><body style="overflow-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><div>Evolution strategies is an alternative to gradient descent in fine-tuning LLMs. Cognizant AI Lab and collaborators recently completed four new papers on it, evaluating it in several new tasks including math and ARC-AGI, extending it to fine-tuning quantized models, demonstrating how LLMs can be trained to know what they know, and showing why high dimensionality of LLMs is a blessing (not a curse) in fine tuning. For more details, see</div><div><br></div><div>Overview blog: https://cgnz.at/6005QZNMb</div><div>Papers: https://cgnz.at/6007QZN13 (tasks)</div><div> https://cgnz.at/6003QZNHN (quantized)</div><div> https://cgnz.at/6002QZNGG (metacognition)</div><div> https://cgnz.at/6000QZNyI (blessing)</div><div>Animated illustration: <a href="https://x.com/RobertoDailey1/status/2026791418380087442">https://x.com/RobertoDailey1/status/2026791418380087442</a></div><div><br></div><div>-- Risto</div><div><br></div></body></html>