SoMark 团队开源 1.3B 文档解析模型 OXR，在 olmOCR-Bench、OmniDocBench 和 DocParsingBench 上排名第一。其 post-training 首次引入 RSI 循环，由 VLM 评估输出、rollout 分歧标记不确定、聚类模型归因并指导下一轮数据生成；单张 RTX PRO 6000 上吞吐达 GLM-OCR 的 3.05 倍，为开源模型中最快。
OXR makes parsing failures part of the training process.
During post-training (OXR is the first document parser to use RSI here), a VLM evaluates its outputs, rollout disagreements flag uncertainty, and a separate clustering model groups the causes;
those findings then guide targeted data generation for the next round, and the 1.3B model that comes out of it sits first on olmOCR-Bench, OmniDocBench and DocParsingBench while running 3.05× GLM-OCR's throughput on a single RTX PRO 6000, the fastest open-source result in the speed comparison.
Worth studying (Github link in comment)
Beyond OCR. Meet OXR. Only 1.3B. Open source. #1 on olmOCR-Bench, OmniDocBench & DocParsingBench🏆 #1 in parsing speed among open-source models⚡ RSI-powered. A first in document parsing. Turn documents into Agent-Friendly data. If OXR helps, give us a star on GitHub! 🌟在 X 查看被引用的帖子
来源：Rohan Paul · x.com