SoMark 开源 1.3B 文档解析模型 OXR，三项基准居首
OXR makes parsing failures part of the training process.
SoMark 开源 1.3B 文档解析模型 OXR，在 olmOCR-Bench、OmniDocBench 和 DocParsingBench 上排名第一，在单张 RTX PRO 6000 上的吞吐为 GLM-OCR 的 3.05 倍，是速度对比中最快的开源结果。OXR 在 post-training 阶段引入 RSI：由 VLM 评估自身输出，rollout 分歧标记不确定性，再由单独的聚类模型归因，据此生成下一轮定向数据。Rohan Paul 称 OXR 是首个在文档解析中使用 RSI 的模型，GitHub 链接见评论。
查看原帖超越 OCR。认识 OXR。 仅 1.3B。开源。 在 olmOCR-Bench、OmniDocBench 和 DocParsingBench 上排名第一🏆 开源模型中解析速度第一⚡ 由 RSI 驱动。文档解析领域的首创。 将文档转化为 Agent-Friendly 数据。 如果 OXR 对你有帮助，请在 GitHub 上给我们点个星！🌟 https://t.co/kQlsq8oIdN
查看原文
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! 🌟 https://t.co/kQlsq8oIdN
正文
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)