AI Paper Atlas

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

AcZeroTS: Active Learning for Zero-shot Tissue Segmentation in Pathology Images

Jiao Tang, Junjie Zhou, Bo Qian, Peng Wan, Yingli Zuo, Wei Shao, Daoqiang Zhang

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Prompted SAM segmentation · 0 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
select samples with the combined uncertainty, prototype and contrast query score instead of baseline active learning criteriaAcZeroTS · TNBC (1 unseen), cycle C3hIoU: 46.0 -> 52.3Improved
combined query strategy versus the second-best method on TNBC with 3 unseen classesAcZeroTS · TNBC (3 unseen), cycle C3hIoU: 33.1 -> 37.7Improved
combined query strategy versus second-best method on HPBCAcZeroTS · HPBC (1 unseen), cycle C3hIoU: 32.8 -> 36.8Improved
combined query strategy versus second-best method on Pascal VOCAcZeroTS · Pascal VOC 2012, cycle C4hIoU: 73.0 -> 77.1Improved
use all three query criteria together instead of a single criterion (US, PS or CS alone) (ablation)AcZeroTS · TNBC and HPBChIoUImproved
include the uncertainty score (US) in the query (ablation)AcZeroTS · TNBC and HPBChIoUImproved
include the prototype score (PS) in the query (ablation)AcZeroTS · TNBC and HPBChIoUImproved
include the contrast score (CS) in the query (ablation)AcZeroTS · TNBC and HPBChIoUImproved
use ProZS prototype-guided model with LLM descriptions and pixel contrastive loss instead of TagCLIPPLIP-based zero-shot segmentation · TNBC (2 unseen)mIoU(U): 30.1 -> 33.7Improved
use ProZS instead of TagCLIP on HPBCPLIP-based zero-shot segmentation · HPBC (1 unseen)mIoU(U): 32.1 -> 35.1Improved

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