AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

HiMTok: Learning Hierarchical Mask Tokens for Image Segmentation with Large Multimodal Model

Tao Wang, Changxu Cheng, Lingfeng Wang, Senda Chen, Wuyue Zhao

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 18 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
add hierarchical mask loss (HML) (ablation)HiMTok-8B · RefCOCO valcIoU: 79.2 -> 81.1Improved
add hierarchical mask loss (HML) (ablation)HiMTok-8B · RefCOCO+ valcIoU: 64.7 -> 77.1Improved
add hierarchical mask loss (HML) (ablation)HiMTok-8B · RefCOCOg valcIoU: 63.9 -> 75.8Improved
increase mask tokens from 16 to 32 (ablation)HiMTok-8B · RefCOCO valcIoU: 82.8% cIoU, additional 2.5% cIoUImproved
increase mask token length before box prediction (ablation)HiMTok-8B · RefCOCO val (REC)Acc@0.9Improved
add box-to-mask information flow (ablation)HiMTok-8B · REScIoUNo clear effect
add finetuning on RefCOCO/+/g training setHiMTok-8B · RefCOCO valcIoU: 81.1 -> 85.0Improved
add finetuned SAM refinement of de-tokenized maskHiMTok-8B (ft) · RefCOCO valcIoU: 85.0 -> 85.9Improved
add Mask Perception training data (ablation)HiMTok-8B · Mask Perception testaccuracy: 63.1 -> 81.8Improved
add joint segmentation training (HiMTok) to InternVL2.5-8BInternVL2.5-8B · RefCOCO val (REC)Acc@0.5: 90.3 -> 92.9Improved
finetune on segmentation-only dataHiMTok-8B · MME countscore: 160 -> 113Hurt
add joint segmentation training (HiMTok) to InternVL2.5-8BInternVL2.5-8B · MME scenescore: 154 -> 163Improved

Later papers that compared with this method

1 later accepted paper reports HiMTok-8B as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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