{"slug": "look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion", "title": "Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs", "summary": "A research paper proposes certificate-guided visual routing for diffusion vision-language models (DVLMs), which process their full input at every denoising step even though visual tokens account for 89-97% of that input on the benchmarks the authors studied. The method targets the repeated re-processing of the same image during decoding, an inefficiency that existing visual token pruning methods only partially relieve.", "body_md": "Diffusion vision-language models (DVLMs) process their entire input at every denoising step. Since visual tokens make up 89-97% of this input on the benchmarks we study, most of the decoding computation is spent repeatedly processing the same image. Visual token pruning methods partially relieve thi", "url": "https://wpnews.pro/news/look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion", "canonical_source": "https://aiflash.com/news/142271/", "published_at": "2026-10-30 06:30:11+00:00", "updated_at": "2026-10-30 06:50:36.258625+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "computer-vision", "ai-research"], "entities": [], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion", "markdown": "https://wpnews.pro/news/look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion.md", "text": "https://wpnews.pro/news/look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion.txt", "jsonld": "https://wpnews.pro/news/look-only-when-needed-certificate-guided-visual-routing-for-efficient-diffusion.jsonld"}}