AI Paper Atlas

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

Rethinking MLLM Itself as a Segmenter with a Single Segmentation Token

Anqi Zhang, Xiaokang Ji, Guangyu Gao, Jianbo Jiao, Chi Harold Liu, Yunchao Wei

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 2 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 uncompressed encoder features F_HQ instead of compressed LLM features (ablation)SELF1E · RefCOCO valcIoU: 76.6 -> 78.7Improved
add Residual Features Refilling on top of F_HQ (ablation)SELF1E · RefCOCO testBcIoU: 75.2 -> 76.3Improved
add pixel-unshuffle with RFR (ablation)SELF1E · RefCOCO valcIoU: 79.0 -> 79.8Improved
add Residual Features Amplifier (shared MLP + separate LLM MLP) (ablation)SELF1E · RefCOCO avg cIoUcIoU: 78.9 -> 80.0Improved
use three independent MLPs in RFA instead of two (ablation)SELF1E · RefCOCO avg cIoUcIoU: 80.0 -> 79.7Hurt
pixel-unshuffle on residual features only (ablation)SELF1E · RefCOCO avg cIoUcIoU: 78.9 -> 79.3No clear effect
enable bidirectional [IMG]->[IMG] attention (ablation)SELF1E · RefCOCOcIoU: 78.4 -> 79.0No clear effect
make [SEG] visible to [IMG] tokens (ablation)SELF1E · RefCOCOcIoU: 78.4 -> 80.0Improved
full bidirectional mask versus [IMG]->[SEG] only (ablation)SELF1E · RefCOCOgcIoU: 77.4 -> 76.9No clear effect
scale SELF1E from 2B to 8BSELF1E · ReasonSeg valgIoU: 59.4 -> 65.9Improved
emphasize segmentation sampling (SEG version)SELF1E-8B · RefCOCOg testcIoU: 79.8 -> 82.8Improved
train without any data from target OVS datasetsSELF1E-2B · Pascal VOC-20 (OVS)mIoU: 89.9 -> 44.8Hurt

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