# Drift-Augmented Scoring: Text-Derived Noise Robustness for Zero-Shot Audio-Language Classification

arXiv: https://arxiv.org/abs/2606.04844 · HTML: https://mlsanity.com/paper/2606.04844/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 23 references that are in the map.
- First posted on arXiv 3 Jun 2026.

## Builds on (most cited first, 18 of 18)

- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [Debiasing Vision-Language Models via Biased Prompts](https://mlsanity.com/paper/2302.00070/) (2023)
- [A multi-device dataset for urban acoustic scene classification](https://mlsanity.com/paper/1807.09840/) (2018)
- [Just Shift It: Test-Time Prototype Shifting for Zero-Shot Generalization with Vision-Language Models](https://mlsanity.com/paper/2403.12952/) (2024)
- [Selective Vision-Language Subspace Projection for Few-shot CLIP](https://mlsanity.com/paper/2407.16977/) (2024)
- [CLIPArTT: Adaptation of CLIP to New Domains at Test Time](https://mlsanity.com/paper/2405.00754/) (2024)
- [ReCLAP: Improving Zero Shot Audio Classification by Describing Sounds](https://mlsanity.com/paper/2409.09213/) (2024)
- [M2D-CLAP: Exploring General-purpose Audio-Language Representations Beyond CLAP](https://mlsanity.com/paper/2503.22104/) (2025)
- [Advancing Test-Time Adaptation in Wild Acoustic Test Settings](https://mlsanity.com/paper/2310.09505/) (2023)
- [PAT: Parameter-Free Audio-Text Aligner to Boost Zero-Shot Audio Classification](https://mlsanity.com/paper/2410.15062/) (2024)
- [TSPE: Task-Specific Prompt Ensemble for Improved Zero-Shot Audio Classification](https://mlsanity.com/paper/2501.00398/) (2024)
- [Domain Adaptation Method and Modality Gap Impact in Audio-Text Models for Prototypical Sound Classification](https://mlsanity.com/paper/2506.04376/) (2025)
- [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)
- [EMO-TTA: Improving Test-Time Adaptation of Audio-Language Models for Speech Emotion Recognition](https://mlsanity.com/paper/2509.25495/) (2025)
- [Leveraging Prediction Entropy for Automatic Prompt Weighting in Zero-Shot Audio-Language Classification](https://mlsanity.com/paper/2601.05011/) (2026)

## Abstract (arXiv)

Contrastive audio-language models such as CLAP enable zero-shot audio classification: a sound is labelled by matching its embedding to text prompt embeddings, with no labelled audio. This matching breaks down under acoustic noise, where accuracy and mAP fall by 12-30 percentage points at 0 dB SNR on standard benchmarks. We propose Drift Augmented Scoring (DAS), a small per-class bonus added to the cosine score. The bonus rewards a class when the noisy audio embedding drifts in the direction that the class's noise-conditioned text prompts predict. It is derived from text alone, computed once and cached, and adds a single inner product per class at inference, with no gradients and no test-time batch. On a LAION CLAP backbone, we compare DAS against the four variants of Acevedo et al.'s concurrent method on UrbanSound8K and the full FSD50K eval set, mixing each clip with urban acoustic scene noise across a range of SNRs. DAS improves the metric on every test condition: by +2.60 to +5.75 accuracy points on UrbanSound8K and +1.50 to +1.74 mAP points on FSD50K.