AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

All in Tokens: Unifying Output Space of Visual Tasks via Soft Token

Jia Ning, Chen Li, Zheng Zhang, Chunyu Wang, Zigang Geng, Qi Dai, Kun He, Han Hu

ICCV 2023 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 65 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
use soft token in both task-solver and detokenizer instead of hard inference (ablation)AiT · NYUv2 depthRMSE: 0.3174 -> 0.3080Improved
use soft token in both task-solver and detokenizer (ablation)AiT · COCO instance segmentationmask mAP: 31.1 -> 33.2Improved
add auxiliary loss on detokenizer output (ablation)AiT with soft token · COCO instance segmentationmask mAP: 33.2 -> 34.2Improved
add auxiliary loss on detokenizer output (ablation)AiT with soft token · COCO instance segmentationbox mAP: 43.6 -> 43.3Hurt
add soft token to parallel decoder (ablation)AiT-P (SwinV2-L) · NYUv2 depthRMSE: 0.282 -> 0.275Improved
add mask augmentation with ratio 0.5 in VQ-VAE training (ablation)AiT · NYUv2 depthRMSE: 0.3105 -> 0.3080Improved
increase mask ratio to 0.7 (ablation)AiT · NYUv2 depthRMSE: 0.3080 -> 0.3225Hurt
increase VQ-VAE codebook size from 64 to 256 (ablation)AiT task-solver · NYUv2 depthRMSE: 0.3090 -> 0.3119No clear effect
increase VQ-VAE width from 1.0x to 2.0x (ablation)AiT task-solver · NYUv2 depthRMSE: 0.3080 -> 0.3124No clear effect
reduce VQ-VAE downsample ratio from 32 to 8 for depth (ablation)AiT · NYUv2 depthRMSE: 0.3080 -> 0.3514Hurt
increase number of residual blocks in VQ-VAE from 2 to 3 or more (ablation)AiT · NYUv2 depthRMSE: 0.3080 -> 0.3123Hurt
replace VQ-VAE tokenizer with interpolation tokenizer (ablation)AiT · COCO instance segmentationmask mAP: 33.2 -> 4.2Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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