SOURCE-LINKED INTELLIGENCE

SupGRPO: Enhancing GRPO with Matching-based Online SFT for Text Spotting

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Text spotting requires both accurate text recognition and precise spatial localization. Current specialised spotters excel at predicting tight bounding boxes in natural scenes, but falter on complex or artistic text, whereas multimodal large language models (MLLMs) possess strong recognition capabilities yet remain weak at localisation. To equip the text spotter with general and powerful recognition capabilities and to maximize its localization ability, we explore two MLLM-based fine-tuning methods: Supervised Fine-Tuning (SFT) and reinforcement learning fine-tuning based on Group Relative Pol

Read original source ↗ Open in workspace

recordType

paper

region

Global