Advancing Speech Understanding in Speech-Aware Language Models with GRPO
This work addresses speech understanding for language models, but it is incremental as it builds on prior GRPO applications to SALLMs by focusing on open-format tasks.
The paper tackles open-format speech understanding tasks like Spoken Question Answering and Automatic Speech Translation by applying Group Relative Policy Optimization (GRPO) with BLEU as a reward signal to Speech-Aware Large Language Models (SALLMs), showing it surpasses standard Supervised Fine-Tuning across several key metrics.
In this paper, we introduce a Group Relative Policy Optimization (GRPO)-based method for training Speech-Aware Large Language Models (SALLMs) on open-format speech understanding tasks, such as Spoken Question Answering and Automatic Speech Translation. SALLMs have proven highly effective for speech understanding tasks. GRPO has recently gained traction for its efficiency in training LLMs, and prior work has explored its application to SALLMs, primarily in multiple-choice tasks. Building on this, we focus on open-format tasks that better reflect the generative abilities of the models. Our approach leverages GRPO with BLEU as the reward signal to optimize SALLMs, and we demonstrate empirically that it surpasses standard SFT across several key metrics. Finally, we explore the potential of incorporating off-policy samples within GRPO for these tasks, highlighting avenues for further improvement and further research.