From Model Patterns to Abstract Semantics in Compositional Zero-Shot Learning

2026-09-14

AI Summary

本文针对组合零样本学习中抽象与具体语义竞争问题，提出了一种基于人类感知过程的CLEAR框架，通过上下文驱动的具体视觉线索激活和重排序预测来改进模型性能。

Abstract

Compositional Zero Shot Learning aims to recognize unseen compositions by recombining learned primitives. Recent methods rely on vision language models and attempt to explicitly model contextual variations of primitives through multiple representations. However, such approaches are limited by fixed variant capacity and competition between abstract and concrete semantics. In this work, we present a new perspective that views primitive variations as the context-driven activation of concrete visual cues rather than independent entities. Based on it, we propose CLEAR, a CLoze-style rEAsoning-based Re-ranking framework inspired by human perceptual processes. CLEAR extracts conditional variants from the primitive candidate set in a coarse-to-fine manner, performs cloze-style reasoning to infer high-level semantics, and re-ranks predictions to correct biases toward salient concrete primitives. Extensive experiments demonstrate that CLEAR consistently improves the Base Model and outperforms state-of-the-art methods on the challenging C-GQA and MIT-States datasets. Code is available at https://github.com/buptLwz/CLEAR.

Problem

• Compositional Zero Shot Learning
• vision language models
• primitive variations
• abstract semantics
• concrete semantics

Innovation

• CLEAR
• Cloze-style reasoning
• context-driven activation
• coarse-to-fine extraction
• re-ranking