AI Paper Atlas

Research map / Continual learning / Class-incremental

Point-UQ: An Uncertainty-Quantification Paradigm for Point Cloud Few-Shot Class Incremental Learning

Xiangqi Li, Libo Huang, Jiarui Zhao, Weilun Feng, Chuanguang Yang, Zhulin An, Yongjun Xu

ICLR 2026 · Continual learning › Class-incremental · approach: Few-shot class-incremental prototypes

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 AAE multi-scale attention fusion (ablation)Point-UQ (with UDD) · ShapeNet to CO3D cross-datasetharmonic accuracy average: 60.0 -> 64.8Improved
add UDD entropy-based decision fusion (ablation)Point-UQ (with AAE) · ShapeNet to CO3D cross-datasetharmonic accuracy average: 55.7 -> 64.8Improved
add both AAE and UDD (ablation)baseline without modules · ShapeNet to CO3D cross-datasetharmonic accuracy average: 54.5 -> 64.8Improved
replace fixed-rule fusion with self-attention fusion (ablation)Deep-Semantic-only · ShapeNet to CO3D cross-datasetharmonic accuracy average: 60.0 -> 64.8Improved
replace LayerWise-To-Last fusion with self-attention fusion (ablation)Point-UQ · ShapeNet to CO3D cross-datasetharmonic accuracy average: 61.0 -> 64.8Improved
replace Symmetric-Cross-Fusion with self-attention fusion (ablation)Point-UQ · ShapeNet to CO3D cross-datasetharmonic accuracy average: 59.4 -> 64.8Improved
use sigmoid instead of softplus_norm activation (ablation)UDD weighting · ShapeNet to CO3D cross-datasetrelative accuracy drop: 18.1 -> 12.1Improved
use sigmoid instead of relu_sigmoid activation (ablation)UDD weighting · ShapeNet to CO3D cross-datasetrelative accuracy drop: 12.8 -> 12.1No clear effect
use entropy instead of semantic confidence gap as uncertainty (ablation)UDD · ShapeNet to CO3D cross-datasetrelative accuracy drop: 13.7 -> 12.1Improved
use entropy instead of semantic-geometric consistency as uncertainty (ablation)UDD · ShapeNet to CO3D cross-datasetrelative accuracy drop: 13.1 -> 12.1Improved
raise lambda from 0.5 to 2.0 (ablation)UDD entropy mapping · ShapeNet to CO3D cross-datasetharmonic accuracy average: 10.9 -> 64.8Improved
raise lambda from 2.0 to 3.0 (ablation)UDD entropy mapping · ShapeNet to CO3D cross-datasetharmonic accuracy average: 64.8 -> 66.5Mixed

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