Transferable Decoding with Visual Entities for Zero-Shot Image Captioning

TL;DR

ViECap introduces an entity-aware decoding approach for zero-shot image captioning, reducing object hallucination and improving cross-domain transferability by guiding language models to focus on actual visual entities.

Contribution

The paper proposes ViECap, a novel transferable decoding model that uses entity-aware hard prompts to enhance zero-shot image captioning across seen and unseen domains.

Findings

Sets new state-of-the-art in cross-domain captioning

Maintains performance when transferring from in-domain to out-of-domain scenarios

Performs competitively in in-domain captioning

Abstract

Image-to-text generation aims to describe images using natural language. Recently, zero-shot image captioning based on pre-trained vision-language models (VLMs) and large language models (LLMs) has made significant progress. However, we have observed and empirically demonstrated that these methods are susceptible to modality bias induced by LLMs and tend to generate descriptions containing objects (entities) that do not actually exist in the image but frequently appear during training (i.e., object hallucination). In this paper, we propose ViECap, a transferable decoding model that leverages entity-aware decoding to generate descriptions in both seen and unseen scenarios. ViECap incorporates entity-aware hard prompts to guide LLMs' attention toward the visual entities present in the image, enabling coherent caption generation across diverse scenes. With entity-aware hard prompts, ViECap is capable of maintaining performance when transferring from in-domain to out-of-domain scenarios. Extensive experiments demonstrate that ViECap sets a new state-of-the-art cross-domain (transferable) captioning and performs competitively in-domain captioning compared to previous VLMs-based zero-shot methods. Our code is available at: https://github.com/FeiElysia/ViECap