Search papers, labs, and topics across Lattice.
This paper introduces PolyOCR, a family of unified OCR foundation models designed to handle recognition, parsing, and reasoning in complex visual scenes, addressing limitations of existing specialized OCR systems. The models leverage a shared instruction-following framework and a novel data engine for quality-verified OCR supervision, alongside Competence-Guided Policy Optimization for effective training. PolyOCR achieves state-of-the-art performance across multiple benchmarks, demonstrating a significant step towards text-centric visual intelligence.
Unified OCR foundation models achieve state-of-the-art visual intelligence by integrating recognition, parsing, and reasoning through instruction-following and a novel competence-guided policy optimization.
Optical Character Recognition (OCR) is evolving from plain-text transcription toward general visual intelligence, requiring models to recognize, localize, and reason over textual information in complex visual environments. However, existing OCR systems often excel at only some tasks and struggle to balance recognition, parsing, and reasoning across scenarios. In this report, we present PolyOCR, a family of unified OCR foundation models of varying scales. PolyOCR combines a shared instruction-following framework with a large-scale data engine that converts heterogeneous visual resources into quality-verified OCR supervision. We introduce Competence-Guided Policy Optimization, which combines verifier-based Group Relative Policy Optimization with on-policy distillation through sample-wise routing based on teacher reliability and the teacher--student competence gap. We also introduce OCRBench v2.1, our revision of OCRBench v2 with manually verified annotation corrections and task-aligned scoring metrics. Extensive experiments across OCRBench v2.1, CC-OCR, in-house KIE Benchmark, OmniDocBench v1.6 and MDPBench demonstrate that PolyOCR achieves state-of-the-art or highly competitive performance.
