arXiv AI
Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jun 11
Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models
arXiv:2606. 12412v1 Announce Type: cross Abstract: Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory.
arXiv AI
Oct 7
Later Is Better: Token Reduction for ViTs Under Distribution Shift
arXiv:2610.07758v1 Announce Type: cross Abstract: Training-free token reduction accelerates vision transformers by removing redundant tokens across layers, recovering most of the original accuracy at...
arXiv Computer Vision
Sep 1
Elastic Token Compression for Pixel-Space Diffusion Transformers
arXiv:2608.29281v1 Announce Type: new Abstract: Natural images concentrate their detail in a small fraction of the frame, yet diffusion models spend a full token on every patch, in every layer and at...
arXiv Computer Vision
2d ago
Rescuing Your Pruning: Rethinking Visual Token Pruning and Post-Training Recovery in MLLMs
arXiv:2610.33688v1 Announce Type: new Abstract: Visual token pruning substantially reduces the inference cost of multimodal large language models (MLLMs), yet existing methods are mostly training-fre...
arXiv AI
Sep 2
Vision Is Not Overhead: One-Pass Block Drafting for Lossless Speculative Decoding in Vision-Language Models
The paper introduces GLANCE, a one‑pass block drafting method that enables lossless speculative decoding for vision‑language models. By using a block‑diffusion head that reads the fused vision‑language state, GLANCE eliminates the need for the drafter to process the image at every step, allowing it to fill an entire block in a single forward pass. Experiments show that GLANCE can decode up to 2.93× faster than autoregressive decoding while maintaining exact greedy decoding results across multiple tasks.
arXiv Machine Learning
Jun 11
Visual-Redundancy-Controlled Parallel Decoding for Diffusion-Based Multimodal Large Language Models
arXiv:2605. 25820v2 Announce Type: replace Abstract: Diffusion-based multimodal large language models (dMLLMs) decode by iteratively predicting tokens at multiple masked positions in parallel.