Classifier-Guided Captioning Across Modalities

TL;DR

This paper presents a novel method for adapting captioning systems to different modalities, such as audio, by guiding a frozen language model with a classifier trained on GPT-4 generated data, enhancing zero-shot performance.

Contribution

Introduces a classifier-guided framework that adapts existing captioning models to new modalities without retraining, using GPT-4 generated data for guidance.

Findings

Improves zero-shot audio captioning performance.

Sets new state-of-the-art results in zero-shot audio captioning.

Operates solely during inference without retraining the captioning model.

Abstract

Most current captioning systems use language models trained on data from specific settings, such as image-based captioning via Amazon Mechanical Turk, limiting their ability to generalize to other modality distributions and contexts. This limitation hinders performance in tasks like audio or video captioning, where different semantic cues are needed. Addressing this challenge is crucial for creating more adaptable and versatile captioning frameworks applicable across diverse real-world contexts. In this work, we introduce a method to adapt captioning networks to the semantics of alternative settings, such as capturing audibility in audio captioning, where it is crucial to describe sounds and their sources. Our framework consists of two main components: (i) a frozen captioning system incorporating a language model (LM), and (ii) a text classifier that guides the captioning system. The classifier is trained on a dataset automatically generated by GPT-4, using tailored prompts specifically designed to enhance key aspects of the generated captions. Importantly, the framework operates solely during inference, eliminating the need for further training of the underlying captioning model. We evaluate the framework on various models and modalities, with a focus on audio captioning, and report promising results. Notably, when combined with an existing zero-shot audio captioning system, our framework improves its quality and sets state-of-the-art performance in zero-shot audio captioning.