AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Parallel Vision Token Scheduling for Fast and Accurate Multimodal LMMs Inference

Wengyi Zhan, Mingbao Lin, Zhihang Lin, Rongrong Ji

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Training-free token reduction for MLLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase migration depth n from 3 to 15 (ablation)ParVTS · LLaVA-1.5-7B, 64 tokensaverage retained (%): 92.91% -> 94.38%Improved
increase migration depth n from 3 to 15 (ablation)ParVTS · LLaVA-1.5-7B, 192 tokensaverage retained (%): 97.95% -> 95.86%Hurt
use parallel scheduling instead of non-subject-first sequential scheduling (ablation)ParVTS · LLaVA-1.5-7B, 64 tokensaverage retained (%): 89.80% -> 92.91%Improved
use parallel scheduling instead of subject-first sequential scheduling (ablation)ParVTS · LLaVA-1.5-7B, 64 tokensaverage retained (%): 85.97% -> 92.91%Improved
use parallel scheduling instead of subject-first sequential scheduling (ablation)ParVTS · LLaVA-1.5-7B, 192 tokensaverage retained (%): 77.77% -> 97.95%Improved
use parallel scheduling instead of non-subject-first sequential scheduling (ablation)ParVTS · LLaVA-1.5-7B, 192 tokensaverage retained (%): 97.28% -> 97.95%Improved
add non-subject path with equal fusion weights alpha=beta=0.5 instead of subject-only beta=1.0 (ablation)ParVTS · LLaVA-1.5-7B, 64 tokensaverage retained (%): 91.55% -> 92.91%Improved
use equal fusion weights instead of alpha=1.0 beta=0.0 (ablation)ParVTS · LLaVA-1.5-7B, 64 tokensaverage retained (%): 92.52% -> 92.91%Improved
use ParVTS instead of SAINT at 64 tokensSAINT · LLaVA-1.5-7B, 64 tokensaverage retained (%): 85.69% -> 92.45%Improved
use ParVTS instead of FastV at -88.89% tokensFastV · Qwen2.5-VL-3Baverage retained (%): 67.46% -> 77.21%Improved
use ParVTS instead of FastV at -88.89% tokensFastV · InternVL2.5-2Baverage retained (%): 91.43% -> 95.88%Improved
use ParVTS instead of FastV at -49.80% tokensFastV · Video-LLaVA-7Baccuracy: 58.1 -> 57.8No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy