AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models

Chen Ju, Haicheng Wang, Haozhe Cheng, Xu Chen, Zhonghua Zhai, Weilin Huang, Jinsong Lan, Shuai Xiao, Bo Zheng

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 6 citations (OpenAlex)

Read the paper

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
add mutual redundancy alone to token merging (ablation)Turbo (BLIP captioning) · COCO captioningCIDEr: 112.8 -> 120.9Improved
add semantic value alone to token merging (ablation)Turbo (BLIP captioning) · COCO captioningCIDEr: 112.8 -> 123.8Improved
combine mutual redundancy and semantic value instead of either alone (ablation)Turbo (BLIP captioning) · COCO captioningCIDEr: 123.8 -> 130.0Improved
use multiplicative instead of additive fusion of the two components (ablation)Turbo · COCO captioningCIDEr: 130.0 -> 129.9No clear effect
use dynamic per-layer fusion (S4) instead of simple weighted difference (ablation)Turbo · COCO captioningCIDEr: +0.3 CIDErNo clear effect
add minimum token-length threshold at large drop ratio (ablation)Turbo · COCO captioning, ratio 65CIDEr: 95.5 -> 115.3Improved
vary balancing coefficient alpha from 1 to 20Turbo · BLIP captioningperformanceNo clear effect
use Turbo instead of UPop pruning for retrieval accelerationBLIP · Flickr30k retrievalimage-to-text R@1: 92.5 -> 96.8Improved
use Turbo instead of ToMe for token mergingViT-B/16 AugReg · ImageNet-1ktop-1 accuracy: 82.6 -> 83.1Improved
apply GPTQ 4-bit quantization to BLIP2BLIP2 · COCO captioningthroughput: 29.5 -> 30.1 (1.02x)No clear effect
stack Turbo on top of UPop pruningBLIP · COCO captioningthroughput speedup: 1.98x -> 2.19xImproved

Later papers that compared with this method

4 later accepted papers report Turbo-8B as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.

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