GSE: Evaluating Sticker Visual Semantic Similarity via a General Sticker Encoder

TL;DR

This paper introduces GSE, a new versatile sticker encoder, and Triple-S, a benchmark dataset, to improve understanding of sticker semantics and support downstream tasks like emotion classification and retrieval.

Contribution

The paper presents the first benchmark for sticker semantic similarity and a novel general sticker encoder that outperforms existing models on unseen stickers.

Findings

GSE achieves superior performance on unseen stickers.

Existing models struggle to capture nuanced sticker semantics.

Triple-S provides a standardized evaluation dataset.

Abstract

Stickers have become a popular form of visual communication, yet understanding their semantic relationships remains challenging due to their highly diverse and symbolic content. In this work, we formally {define the Sticker Semantic Similarity task} and introduce {Triple-S}, the first benchmark for this task, consisting of 905 human-annotated positive and negative sticker pairs. Through extensive evaluation, we show that existing pretrained vision and multimodal models struggle to capture nuanced sticker semantics. To address this, we propose the {General Sticker Encoder (GSE)}, a lightweight and versatile model that learns robust sticker embeddings using both Triple-S and additional datasets. GSE achieves superior performance on unseen stickers, and demonstrates strong results on downstream tasks such as emotion classification and sticker-to-sticker retrieval. By releasing both Triple-S and GSE, we provide standardized evaluation tools and robust embeddings, enabling future research in sticker understanding, retrieval, and multimodal content generation. The Triple-S benchmark and GSE have been publicly released and are available here.