Making Better Mistakes in CLIP-Based Zero-Shot Classification with Hierarchy-Aware Language Prompts
This work addresses error reduction in zero-shot classification for computer vision applications, but it is incremental as it builds on existing CLIP and LLM methods.
The paper tackles the problem of improving mistake severity in CLIP-based zero-shot image classification by using hierarchy-aware language prompts, achieving better performance across five datasets with varying label hierarchies.
Recent studies are leveraging advancements in large language models (LLMs) trained on extensive internet-crawled text data to generate textual descriptions of downstream classes in CLIP-based zero-shot image classification. While most of these approaches aim at improving accuracy, our work focuses on ``making better mistakes", of which the mistakes' severities are derived from the given label hierarchy of downstream tasks. Since CLIP's image encoder is trained with language supervising signals, it implicitly captures the hierarchical semantic relationships between different classes. This motivates our goal of making better mistakes in zero-shot classification, a task for which CLIP is naturally well-suited. Our approach (HAPrompts) queries the language model to produce textual representations for given classes as zero-shot classifiers of CLIP to perform image classification on downstream tasks. To our knowledge, this is the first work to introduce making better mistakes in CLIP-based zero-shot classification. Our approach outperforms the related methods in a holistic comparison across five datasets of varying scales with label hierarchies of different heights in our experiments. Our code and LLM-generated image prompts: \href{https://github.com/ltong1130ztr/HAPrompts}{https://github.com/ltong1130ztr/HAPrompts}.