| --- |
| base_model: unsloth/Qwen3-VL-2B-Instruct |
| library_name: peft |
| license: apache-2.0 |
| pipeline_tag: image-text-to-text |
| tags: |
| - arabic |
| - ocr |
| - qwen3-vl |
| - lora |
| - unsloth |
| - document-ai |
| --- |
| |
| # Alhazen-OCR |
|
|
| QLoRA adapter on [`unsloth/Qwen3-VL-2B-Instruct`](https://huggingface.co/unsloth/Qwen3-VL-2B-Instruct) for Arabic institutional OCR: printed forms, invoices, and handwriting-heavy pages. |
|
|
| Named after Ibn al-Haytham (Alhazen), the 11th-century scholar who founded the science of optics. |
|
|
| - **Base:** Qwen3-VL-2B-Instruct (Apache-2.0) |
| - **Training data:** [`context212/context212-alhazen-ocr`](https://huggingface.co/datasets/context212/context212-alhazen-ocr) (khattmix rebuild) |
| - **Code:** [`github.com/context212/atlas-ocr`](https://github.com/context212/atlas-ocr) |
| - **Experiment sibling:** [`context212/alhazen-ocr-khattmix`](https://huggingface.co/context212/alhazen-ocr-khattmix) |
|
|
| ## Results |
|
|
| CER / WER (lower is better), greedy decoding, **50 samples** each on the held-out eval split and on external [`ahmedheakl/arocrbench_khatt`](https://huggingface.co/datasets/ahmedheakl/arocrbench_khatt): |
|
|
| | Model | Eval CER ↓ | Eval WER ↓ | KHATT CER ↓ | KHATT WER ↓ | |
| |---|---:|---:|---:|---:| |
| | Qwen3-VL-2B-Instruct (base) | 0.792 | 0.947 | 1.893 | 1.801 | |
| | **Alhazen-OCR** | **0.328** | **0.504** | **0.841** | **1.193** | |
|
|
| Held-out character error: **0.79 → 0.33**. External KHATT handwriting: **1.89 → 0.84** CER. |
|
|
| This release (khattmix) adds line-level KHATT handwriting (~30% of the mix) on top of synthetic printed, invoices, and paragraph handwriting; KHATT-bench transcript overlap is removed from train. |
|
|
| ## Training |
|
|
| - **Objective:** QLoRA SFT, one epoch |
| - **LR:** 2e-5 cosine, warmup 5% |
| - **Effective batch:** 16 (8 × grad accum 2) on A100 80GB |
| - **LoRA:** r=16, RSLoRA, vision + language layers |
| - **Mix:** ~30% loay synthetic / 15% invoices / 30% KHATT lines / 23% KHATT paragraphs / 2% historical (~18k train rows) |
|
|
| ## Usage |
|
|
| Serve the base with the adapter attached (vLLM example): |
|
|
| ```bash |
| vllm serve Qwen/Qwen3-VL-2B-Instruct \ |
| --enable-lora \ |
| --lora-modules alhazen=context212/alhazen-ocr \ |
| --limit-mm-per-prompt '{"image": 1}' \ |
| --mm-processor-cache-gb 0 \ |
| --no-enable-prefix-caching |
| ``` |
|
|
| Prompt: |
|
|
| > Extract all the text from this image, preserving the original reading order. |
|
|
| Keep `temperature=0` for transcription. |
|
|
| ## Limitations |
|
|
| - Tuned for Arabic institutional paperwork; not a general multilingual OCR. |
| - KHATT remains harder than printed forms — scores above 1.0 CER mean many lines are still wrong. |
| - Eval above is capped at 50 samples per split; treat as directional until a full-split rerun. |
|
|
| ## Citation |
|
|
| ```bibtex |
| @misc{alhazen-ocr-2026, |
| title = {Alhazen-OCR}, |
| author = {Context212}, |
| year = {2026}, |
| howpublished = {\url{https://huggingface.co/context212/alhazen-ocr}} |
| } |
| ``` |
|
|
