Towards Reliable Large Audio Language Model

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ma, Ziyang, Li, Xiquan, Song, Yakun, Chen, Wenxi, Du, Chenpeng, Wu, Jian, Chen, Yuanzhe, Chen, Zhuo, Wang, Yuping, Wang, Yuxuan, Chen, Xie
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918034381209600
author Ma, Ziyang
Li, Xiquan
Song, Yakun
Chen, Wenxi
Du, Chenpeng
Wu, Jian
Chen, Yuanzhe
Chen, Zhuo
Wang, Yuping
Wang, Yuxuan
Chen, Xie
author_facet Ma, Ziyang
Li, Xiquan
Song, Yakun
Chen, Wenxi
Du, Chenpeng
Wu, Jian
Chen, Yuanzhe
Chen, Zhuo
Wang, Yuping
Wang, Yuxuan
Chen, Xie
contents Recent advancements in large audio language models (LALMs) have demonstrated impressive results and promising prospects in universal understanding and reasoning across speech, music, and general sound. However, these models still lack the ability to recognize their knowledge boundaries and refuse to answer questions they don't know proactively. While there have been successful attempts to enhance the reliability of LLMs, reliable LALMs remain largely unexplored. In this paper, we systematically investigate various approaches towards reliable LALMs, including training-free methods such as multi-modal chain-of-thought (MCoT), and training-based methods such as supervised fine-tuning (SFT). Besides, we identify the limitations of previous evaluation metrics and propose a new metric, the Reliability Gain Index (RGI), to assess the effectiveness of different reliable methods. Our findings suggest that both training-free and training-based methods enhance the reliability of LALMs to different extents. Moreover, we find that awareness of reliability is a "meta ability", which can be transferred across different audio modalities, although significant structural and content differences exist among sound, music, and speech.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Reliable Large Audio Language Model
Ma, Ziyang
Li, Xiquan
Song, Yakun
Chen, Wenxi
Du, Chenpeng
Wu, Jian
Chen, Yuanzhe
Chen, Zhuo
Wang, Yuping
Wang, Yuxuan
Chen, Xie
Sound
Computation and Language
Human-Computer Interaction
Multimedia
Audio and Speech Processing
Recent advancements in large audio language models (LALMs) have demonstrated impressive results and promising prospects in universal understanding and reasoning across speech, music, and general sound. However, these models still lack the ability to recognize their knowledge boundaries and refuse to answer questions they don't know proactively. While there have been successful attempts to enhance the reliability of LLMs, reliable LALMs remain largely unexplored. In this paper, we systematically investigate various approaches towards reliable LALMs, including training-free methods such as multi-modal chain-of-thought (MCoT), and training-based methods such as supervised fine-tuning (SFT). Besides, we identify the limitations of previous evaluation metrics and propose a new metric, the Reliability Gain Index (RGI), to assess the effectiveness of different reliable methods. Our findings suggest that both training-free and training-based methods enhance the reliability of LALMs to different extents. Moreover, we find that awareness of reliability is a "meta ability", which can be transferred across different audio modalities, although significant structural and content differences exist among sound, music, and speech.
title Towards Reliable Large Audio Language Model
topic Sound
Computation and Language
Human-Computer Interaction
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2505.19294