OXR 把解析失败纳入训练流程，1.3B 模型登顶三份 OCR 基准
文档解析模型 OXR 把解析失败直接喂回训练循环，1.3B 参数就拿下 olmOCR-Bench 等三份基准第一，速度还是 GLM-OCR 的 3.05 倍。
OXR 在后训练阶段首次引入 RSI，让 VLM 评估自身输出，用 rollout 分歧标记不确定样本，再由聚类模型归因失败原因。这些发现用于指导下一轮的针对性数据生成。产出的 1.3B 模型在 olmOCR-Bench、OmniDocBench 和 DocParsingBench 上排名第一。单张 RTX PRO 6000 上的吞吐达 GLM-OCR 的 3.05 倍，是该速度对比中最快的开源结果。
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)
