https://yutianyu.cn/tag/post-training/Post-trainingWowchemy (https://wowchemy.com)en-usThu, 30 Apr 2026 00:00:00 +0000https://yutianyu.cn/media/icon_hu0ee3417c3693759e075e80aecf96b070_11607_512x512_fill_lanczos_center_3.pnghttps://yutianyu.cn/tag/post-training/Thu, 30 Apr 2026 00:00:00 +0000https://yutianyu.cn/project/minicpm-v-o/Thu, 01 Jan 2026 00:00:00 +0000https://yutianyu.cn/project/minicpm-v-o/<p>Earlier releases include <a href="https://github.com/OpenBMB/MiniCPM-V/blob/main/docs/minicpm_v4_en.md" target="_blank" rel="noopener">MiniCPM-V 4.0</a>, <a href="https://github.com/OpenBMB/MiniCPM-V/blob/main/docs/minicpm_o2dot6_en.md" target="_blank" rel="noopener">MiniCPM-o 2.6</a>, and <a href="https://github.com/OpenBMB/MiniCPM-V/blob/main/docs/minicpm_v2dot6_en.md" target="_blank" rel="noopener">MiniCPM-V 2.6</a>. See the <a href="https://github.com/OpenBMB/MiniCPM-V/blob/main/docs/project_en.md" target="_blank" rel="noopener">official model archive</a> for more releases.</p> <p>I am a core contributor to the <a href="https://github.com/OpenBMB/MiniCPM-V" target="_blank" rel="noopener">MiniCPM-V/o</a> multimodal model family, covering post-training, evaluation, release, and iteration. I was the first student author of <a href="https://github.com/OpenBMB/MiniCPM-V" target="_blank" rel="noopener">MiniCPM-V</a> and the first author of <a href="https://arxiv.org/abs/2509.18154" target="_blank" rel="noopener">MiniCPM-V 4.5</a>.</p> <p>My work focuses on post-training recipes that make efficient MLLMs strong in multimodal understanding, document intelligence, and real deployment scenarios.</p> <p>See <a href="https://yutianyu.cn/project/minicpm-v-4-7/">MiniCPM-V 4.7</a> for the latest vision-language release, and <a href="https://yutianyu.cn/project/minicpm-ocr/">MiniCPM-OCR</a> for document parsing and OCR quality assessment.</p>Tue, 16 Sep 2025 00:00:00 +0000