AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Identifying Latent Concepts and Structures for Generalized Category Discovery

Boyang Dai, Chaoqi Chen, Yizhou Yu

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 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 CPF-GCD moduleSelEx · CUB-200ACC All (%): 75.6 -> 79.8Improved
add CPF-GCD moduleLegoGCD · Fine-grained novel classesACC New (%): 8.1%Improved
add CPF-GCD moduleSimGCD · Fine-grained averageACC All (%): 56.8 -> 59.1Improved
add CPF-GCD moduleSelEx · ImageNet-100ACC New (%): 77.7 -> 80.8Improved
add CPF-GCD moduleLegoGCD · CIFAR-10ACC Old (%): 96.1 -> 95.8Hurt
add CPF-GCD moduleSimGCD · CIFAR-100ACC All (%): 80.1 -> 80.9Improved
use refined primitive aggregation instead of global patch mean (ablation)SimGCD + CPF-GCD · CUB-200ACC New (%): 60.3 -> 63.1Improved
add data-assignment (ablation)SimGCD + CPF-GCD · CUB-200ACC All (%): 62.2 -> 63.9Improved
add prior-assignment (ablation)SimGCD + CPF-GCD · Stanford-CarsACC New (%): 48.9 -> 49.8Improved
add learnable fusion gates alpha (ablation)SimGCD + CPF-GCD · CUB-200ACC All (%): 63.2 -> 63.9Improved
vary number of primitives M (ablation)CPF-GCD · fine-grained benchmarksACCNo clear effect
add CPF-GCD module to CMSCMS · CUB-200category number estimation error (%): 18 -> 11.5Improved

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