A two-stage framework that combines self-guided pseudo-labeling with neighborhood-based representation balancing sets new state-of-the-art accuracy on long-tailed generalized category discovery benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Long-Tailed Learning for Generalized Category Discovery
A two-stage framework that combines self-guided pseudo-labeling with neighborhood-based representation balancing sets new state-of-the-art accuracy on long-tailed generalized category discovery benchmarks.