AI Paper Atlas

Research map / Continual learning / Class-incremental

Virtual Category-Guided Continual Generalized Category Discovery

Jiahui Xiong, Qiuxia Lai, Hongsong Wang

ECCV 2026 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 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
add virtual category learning (VCL) (ablation)ENCL-less baseline · CIFAR-100avg accuracy (All): 69.00 -> 70.04Improved
add virtual category learning (VCL) (ablation)baseline · Tiny-ImageNetavg accuracy (All): 63.22 -> 63.13No clear effect
add expanded neighborhood contrastive learning (ENCL) (ablation)baseline · CIFAR-100avg accuracy (All): 69.00 -> 69.57Improved
add expanded neighborhood contrastive learning (ENCL) (ablation)baseline · Tiny-ImageNetavg accuracy (All): 63.22 -> 64.22Improved
add ENCL on top of VCL (ablation)VCL · CIFAR-100avg accuracy (All): 70.04 -> 71.27Improved
add VCL on top of ENCL (ablation)ENCL · Tiny-ImageNetavg accuracy (All): 64.22 -> 64.40Improved
add VCL on top of ENCL (ablation)ENCL · CIFAR-100New accuracy: 51.10 -> 56.61Improved
ambiguity threshold 0.15 instead of 0.07 (ablation)VCL · CIFAR-100avg accuracy: 68.63 -> 71.27Improved
ambiguity threshold 0.15 instead of 0.3 (ablation)VCL · CIFAR-100avg accuracy: 68.42 -> 71.27Improved
neighborhood size 3 instead of 4 (ablation)ENCL · CIFAR-100avg accuracy: 69.09 -> 71.27Improved
neighborhood size 3 instead of 2 (ablation)ENCL · CIFAR-100avg accuracy: 70.52 -> 71.27Improved
add ENCL (training cost) (ablation)model without ENCL · CIFAR-100training time s/epoch: 338 -> 384Hurt

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