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

arXiv: https://arxiv.org/abs/2609.07081 · HTML: https://mlsanity.com/paper/2609.07081/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 38 of its 46 references that are in the map.
- First posted on arXiv 7 Sep 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 30 of 38)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model](https://mlsanity.com/paper/2504.07615/) (2025)
- [mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding](https://mlsanity.com/paper/2403.12895/) (2024)
- [General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model](https://mlsanity.com/paper/2409.01704/) (2024)
- [TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document](https://mlsanity.com/paper/2403.04473/) (2024)
- [Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes](https://mlsanity.com/paper/1908.08207/) (2019)
- [ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network](https://mlsanity.com/paper/2002.10200/) (2020)
- [ICDAR 2019 Robust Reading Challenge on Reading Chinese Text on Signboard](https://mlsanity.com/paper/1912.09641/) (2019)
- [Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting](https://mlsanity.com/paper/2007.09482/) (2020)
- [FOTS: Fast Oriented Text Spotting with a Unified Network](https://mlsanity.com/paper/1801.01671/) (2018)
- [PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model](https://mlsanity.com/paper/2510.14528/) (2025)
- [ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting](https://mlsanity.com/paper/2105.03620/) (2021)
- [DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting](https://mlsanity.com/paper/2211.10772/) (2022)
- [UniDoc: A Universal Large Multimodal Model for Simultaneous Text Detection, Recognition, Spotting and Understanding](https://mlsanity.com/paper/2308.11592/) (2023)
- [SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition](https://mlsanity.com/paper/2203.10209/) (2022)
- [Text Spotting Transformers](https://mlsanity.com/paper/2204.01918/) (2022)
- [Towards End-to-end Text Spotting with Convolutional Recurrent Neural Networks](https://mlsanity.com/paper/1707.03985/) (2017)
- [Text Perceptron: Towards End-to-End Arbitrary-Shaped Text Spotting](https://mlsanity.com/paper/2002.06820/) (2020)
- [PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network](https://mlsanity.com/paper/2104.05458/) (2021)
- [Rethinking Text Segmentation: A Novel Dataset and A Text-Specific Refinement Approach](https://mlsanity.com/paper/2011.14021/) (2020)
- [SPTS: Single-Point Text Spotting](https://mlsanity.com/paper/2112.07917/) (2021)
- [Towards Weakly-Supervised Text Spotting using a Multi-Task Transformer](https://mlsanity.com/paper/2202.05508/) (2022)
- [GLASS: Global to Local Attention for Scene-Text Spotting](https://mlsanity.com/paper/2208.03364/) (2022)

## Abstract (arXiv)

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 Policy Optimisation (GRPO). An interesting finding is that SFT is less effective than GRPO at enhancing recognition, while GRPO is less effective than SFT at enhancing detection. To compensate for each other's shortcomings, we introduce a joint training strategy, SupGRPO, which simultaneously optimizes the model using both SFT and GRPO. SupGRPO employs the specially designed reward functions and develops a matching-based online SFT applied solely to coordinate tokens. It both mitigates the reward sparsity problem of GRPO and avoids the instance order dependency problem of SFT. To evaluate particularly challenging cases, we curate ATS, a dataset for artistic text spotting. Experiments demonstrate that SupGRPO improves both text recognition and detection, and attains superior performance. Our code and dataset will be released at https://github.com/Psycho-9/SupGRPO.