AI Paper Atlas

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

Refer to Any Segmentation Mask Group With Vision-Language Prompts

Shengcao Cao, Zijun Wei, Jason Kuen, Kangning Liu, Lingzhi Zhang, Jiuxiang Gu, HyunJoon Jung, Liang-Yan Gui, Yu-Xiong Wang

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 6 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 non-autoregressive decoding instead of autoregressive (ablation)RAS · ORES MaskGroups-HQcIoU: 45.34 -> 53.75Improved
use non-autoregressive decoding instead of autoregressive (ablation)RAS · ORES MaskGroups-HQlatency: 2.13 -> 0.56Improved
ensemble four visual encoders instead of CLIP only (ablation)RAS · ORES MaskGroups-HQoverall cIoU: 52.44 -> 53.75Improved
ensemble four visual encoders instead of DINOv2 only (ablation)RAS · ORES MaskGroups-HQoverall cIoU: 47.71 -> 53.75Improved
ensemble four visual encoders instead of ConvCLIP only (ablation)RAS · ORES MaskGroups-HQoverall cIoU: 53.53 -> 53.75Improved
use different special tokens for candidate and reference masks instead of none (ablation)RAS · ORES MaskGroups-HQoverall cIoU: 50.13 -> 53.75Improved
use different special tokens instead of the same token (ablation)RAS · ORES MaskGroups-HQoverall cIoU: 48.49 -> 53.75Improved
scale LLM from 7B to 13B (ablation)RAS · ORES overall cIoUcIoU: 52.19 -> 53.93Improved
scale LLM from 7B to 13B (ablation)RAS · RES averagecIoU: 73.7 -> 75.0Improved
finetune on MaskGroups-HQ (ORES-FT)RAS · ORES MaskGroups-HQ overallcIoU: 53.93 -> 73.13Improved
finetune on RES dataRAS · RES averagecIoU: 75.0 -> 77.8Improved
finetune on GRES dataRAS · gRefCOCO averagecIoU: 67.78 -> 71.79Improved

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