https://derekwang2002.github.io/ai-intelligence/en/projects/oppd/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261005-05/ ev-20261005-05-published Tue, 06 Oct 2026 16:01:24 GMT New TRIAL <p>On-policy power distillation (OPPD) trains a model to approximate a sharpened sequence-level answer distribution in one generation. A frozen teacher weights candidates from sequential Monte Carlo, and those weights supervise the student without reference answers. The authors report gains up to 23 points on MATH-500 and 27.3 on GSM8K over the base checkpoint, plus improvements beyond matched-budget GRPO. Code is public.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://arxiv.org/abs/2610.06804">Sharpen Without Search: OPPD</a> · <a href="https://github.com/ArminAzizi98/OPPD">ArminAzizi98/OPPD</a></p>