Author name: Baoyou Chen, Hanchen Xia, Peng Tu, Haojun Shi, Shan Mu, Weihao Yuan, Siyu Zhu

cs.CV, cs.LG

BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation

/

arXiv:2604.16514v2 Announce Type: replace-cross
Abstract: Autoregressive vision-language models (VLMs) deliver strong multimodal capability, but their token-by-token decoding imposes a fundamental inference bottleneck. Diffusion VLMs offer a more para…

cs.CV, cs.LG

BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation

/

arXiv:2604.16514v1 Announce Type: new
Abstract: Autoregressive vision-language models (VLMs) deliver strong multimodal capability, but their token-by-token decoding imposes a fundamental inference bottleneck. Diffusion VLMs offer a more parallel decod…

Scroll to Top