ArmorOCR

ArmorOCR is a two-stage framework for grounded adversarial OCR perception built on Qwen3-VL-8B-Instruct. It enables single-pass inference on the original image, without any inference-time visual transformations or tool assistance.

📖 For training details, the AdvSpot benchmark, and evaluation scripts, please visit the GitHub repo.

Quickstart

pip install transformers==4.57.1 accelerate

from transformers import Qwen3VLForConditionalGeneration, AutoProcessor

model = Qwen3VLForConditionalGeneration.from_pretrained(
    "inclusionAI/ArmorOCR", dtype="auto", device_map="auto",
)
processor = AutoProcessor.from_pretrained("inclusionAI/ArmorOCR")

messages = [{
    "role": "user",
    "content": [
        {"type": "image", "image": "path/to/image.png"},
        {"type": "text",
         "text": ("Please identify the text in the image. "
                  "Put your reasoning inside <analyze></analyze> "
                  "and your final recognized text inside <answer></answer>.")},
    ],
}]

inputs = processor.apply_chat_template(
    messages, tokenize=True, add_generation_prompt=True,
    return_dict=True, return_tensors="pt",
).to(model.device)

out = model.generate(**inputs, max_new_tokens=256)
trimmed = [o[len(i):] for i, o in zip(inputs.input_ids, out)]
print(processor.batch_decode(trimmed, skip_special_tokens=True))

License

Released under the Apache License 2.0. Use is additionally subject to the license and acceptable-use policy of the base model Qwen/Qwen3-VL-8B-Instruct.

Citation

@misc{cao2026armorocrgroundedadversarialvisual,
      title={ArmorOCR: Grounded Adversarial Visual Perception via Observation-Transferred Self-Distillation}, 
      author={Linhan Cao and Siyuan Li and Jun Lan and Liangbo He and Guannan Li and Xiaolei Huang and Jun Jia and Shuheng Zhou and Huijia Zhu and Weiqiang Wang and Wei Sun},
      year={2026},
      eprint={2608.20122},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2608.20122}, 
}