Dissecting Generalized Category Discovery: Multiplex Consensus under Self-Deconstruction

TL;DR

ConGCD introduces a human-inspired approach to generalized category discovery by decomposing objects into primitives and establishing consensus through semantic reconstruction and dynamic scheduling, improving recognition of known and novel categories.

Contribution

This paper presents ConGCD, a novel method that uses primitive decomposition and multiplex consensus to enhance generalized category discovery beyond existing objective function optimization.

Findings

Outperforms existing GCD methods on multiple benchmarks.

Effectively captures class-discriminative and invariant features.

Demonstrates robustness in coarse- and fine-grained recognition tasks.

Abstract

Human perceptual systems excel at inducing and recognizing objects across both known and novel categories, a capability far beyond current machine learning frameworks. While generalized category discovery (GCD) aims to bridge this gap, existing methods predominantly focus on optimizing objective functions. We present an orthogonal solution, inspired by the human cognitive process for novel object understanding: decomposing objects into visual primitives and establishing cross-knowledge comparisons. We propose ConGCD, which establishes primitive-oriented representations through high-level semantic reconstruction, binding intra-class shared attributes via deconstruction. Mirroring human preference diversity in visual processing, where distinct individuals leverage dominant or contextual cues, we implement dominant and contextual consensus units to capture class-discriminative patterns and inherent distributional invariants, respectively. A consensus scheduler dynamically optimizes activation pathways, with final predictions emerging through multiplex consensus integration. Extensive evaluations across coarse- and fine-grained benchmarks demonstrate ConGCD's effectiveness as a consensus-aware paradigm. Code is available at github.com/lytang63/ConGCD.