InternLM 开源 VSR：像素引导的图表解析视觉自精炼框架（ICLR 2026）
GitHub - InternLM/VSR: [ICLR 2026] Official repository of "Visual Self-Refine: A Pixel-Guided Paradigm for Accurate Chart Parsing" · GitHub
InternLM 发布 VSR（Visual Self-Refine），将图表解析变成像素引导的反馈闭环：先预测锚点、渲染锚点、检查渲染后的图表，再依据验证过的像素精炼最终解析结果。该仓库为 ICLR 2026 论文《Visual Self-Refine: A Pixel-Guided Paradigm for Accurate Chart Parsing》的官方实现，采用 Apache-2.0 许可，目前获 35 个 Star。
正文
当前显示原文，中文全文译文尚未就绪。
InternLM; VSR
Repository navigation
NotificationsYou must be signed in to change notification settings; Fork0 (0); Star35 (35)You must be signed in to star a repository
Repository files navigation
README; Apache-2.0 license
Visual Self-Refine (VSR) turns chart parsing into a pixel-guided feedback loop: predict anchors, render them, inspect the rendered chart, and refine the final parsing from the verified pixels.