An Evaluation of Interleaved Instruction Tuning on Semantic Reasoning Performance in an Audio MLLM

TL;DR

This paper investigates how interleaved instruction tuning affects semantic reasoning in an audio multi-modal large language model, showing improvements in reasoning tasks but potential trade-offs in audio labeling ability.

Contribution

It introduces interleaved instruction tuning for audio MLLMs and evaluates its impact on semantic reasoning performance using a new benchmark dataset.

Findings

Zero-shot interleaved prompting improves reasoning performance.

Fine-tuning enhances reasoning further.

Fine-tuning reduces audio labeling accuracy.

Abstract

Standard training for Multi-modal Large Language Models (MLLMs) involves concatenating non-textual information, like vision or audio, with a text prompt. This approach may not encourage deep integration of modalities, limiting the model's ability to leverage the core language model's reasoning capabilities. This work examined the impact of interleaved instruction tuning in an audio MLLM, where audio tokens are interleaved within the prompt. Using the Listen, Think, and Understand (LTU) model as a testbed, we conduct an experiment using the Synonym and Hypernym Audio Reasoning Dataset (SHARD), our newly created reasoning benchmark for audio-based semantic reasoning focusing on synonym and hypernym recognition. Our findings show that while even zero-shot interleaved prompting improves performance on our reasoning tasks, a small amount of fine-tuning using interleaved training prompts improves the results further, however, at the expense of the MLLM's audio labeling ability.