# Adapting Language-Audio Models as Few-Shot Audio Learners

arXiv: https://arxiv.org/abs/2305.17719 · HTML: https://mlsanity.com/paper/2305.17719/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 May 2023.

## Built on by (most cited first, 10 of 10)

- [Separate Anything You Describe](https://mlsanity.com/paper/2308.05037/) (2023)
- [Bridging Paintings and Music -- Exploring Emotion based Music Generation through Paintings](https://mlsanity.com/paper/2409.07827/) (2024)
- [TSPE: Task-Specific Prompt Ensemble for Improved Zero-Shot Audio Classification](https://mlsanity.com/paper/2501.00398/) (2024)
- [Multiple Consistency-guided Test-Time Adaptation for Contrastive Audio-Language Models with Unlabeled Audio](https://mlsanity.com/paper/2412.17306/) (2024)
- [CLAP-S: Support Set Based Adaptation for Downstream Fiber-optic Acoustic Recognition](https://mlsanity.com/paper/2501.09877/) (2025)
- [SPANER: Shared Prompt Aligner for Multimodal Semantic Representation](https://mlsanity.com/paper/2508.13387/) (2025)
- [EMO-TTA: Improving Test-Time Adaptation of Audio-Language Models for Speech Emotion Recognition](https://mlsanity.com/paper/2509.25495/) (2025)
- [A Comprehensive Survey of Challenges and Opportunities of Few-Shot Learning Across Multiple Domains](https://mlsanity.com/paper/2504.04017/) (2025)
- [EmoHeal: An End-to-End System for Personalized Therapeutic Music Retrieval from Fine-grained Emotions](https://mlsanity.com/paper/2509.15986/) (2025)
- [Prototype-Rectified Iterative Self-supervised Manifold Denoising under Severe Acoustic Shift](https://mlsanity.com/paper/2608.15037/) (2026)

## Abstract (arXiv)

We presented the Treff adapter, a training-efficient adapter for CLAP, to boost zero-shot classification performance by making use of a small set of labelled data. Specifically, we designed CALM to retrieve the probability distribution of text-audio clips over classes using a set of audio-label pairs and combined it with CLAP's zero-shot classification results. Furthermore, we designed a training-free version of the Treff adapter by using CALM as a cosine similarity measure. Experiments showed that the proposed Treff adapter is comparable and even better than fully-supervised methods and adaptation methods in low-shot and data-abundant scenarios. While the Treff adapter shows that combining large-scale pretraining and rapid learning of domain-specific knowledge is non-trivial for obtaining generic representations for few-shot learning, it is still limited to audio classification tasks. In the future, we will explore how to use audio-language models in diverse audio domains.