Add model card for BARD-VL

README.md

@@ -0,0 +1,46 @@
+---
+pipeline_tag: image-text-to-text
+---
+
+# BARD-VL
+
+BARD (Bridging AutoRegressive and Diffusion) is a framework that converts a pretrained autoregressive vision-language model (VLM) into a decoding-efficient diffusion VLM (dVLM). By using progressive supervised block merging and stage-wise distillation, BARD-VL achieves significant decoding throughput speedups (up to 3$\times$) compared to the source autoregressive models while maintaining high quality in multimodal tasks.
+
+- **Paper:** [BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation](https://huggingface.co/papers/2604.16514)
+- **Repository:** [https://github.com/fudan-generative-vision/Bard-VL](https://github.com/fudan-generative-vision/Bard-VL)
+- **Project Page:** [https://fudan-generative-vision.github.io/Bard-VL](https://fudan-generative-vision.github.io/Bard-VL)
+
+## Method Overview
+
+The BARD framework introduces two main stages to bridge the gap between autoregressive and diffusion paradigms:
+1. **Progressive Block Merging (PBM):** Gradually enlarges the decoding block size.
+2. **Stage-Wise Distillation (SWD):** Intra-dVLM distillation from a fixed small-block diffusion anchor to recover performance lost at larger blocks.
+
+## Usage
+
+To use BARD-VL, please clone the [official repository](https://github.com/fudan-generative-vision/Bard-VL) and follow the installation instructions. You can then run inference for image and video understanding using the provided `inference.py` script:
+
+```bash
+python3 inference.py \
+  --model_id <path_to_model_checkpoint> \
+  --block_size 4 \
+  --denoising_steps 4 \
+  --confidence_threshold 0.6
+```
+
+## Citation
+
+If you find BARD-VL useful in your research, please cite the following paper:
+
+```bibtex
+@article{chen2026bard,
+  title={BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation},
+  author={Chen, Baoyou and Xia, Hanchen and Tu, Peng and Shi, Haojun and Mu, Shan and Yuan, Weihao and Zhu, Siyu},
+  journal={arXiv preprint arXiv:2604.16514},
+  year={2026}
+}
+```
+
+## Acknowledgements
+
+This project builds on top of [NVIDIA NeMo AutoModel](https://github.com/NVIDIA-NeMo/Automodel).