# Activation Steering for Accent Adaptation in Large Audio Language Models

Source: [Activation Steering for Accent Adaptation in Large Audio Language Models](<https://www.sota2.com/research/paper/69ae53fd2d603e6de3e42afc>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-19T05:00:05.274Z

Authors: Jinuo Sun, Yang Xiao, Sung Kyun Chung, Qiuchi Hu, Gongping Huang, Eun-Jung Holden, Ting Dang

Published: 2026-03-06T01:54:28.000Z

## Abstract

Accent variability remains a major source of errors in automatic speech recognition, yet most adaptation methods rely on parameter fine-tuning without understanding where accent information is encoded. We treat accent variation as an interpretable subspace in hidden representations and investigate whether it can be identified and controlled directly in activation space. We extract layer-wise encoder activations and estimate mean-shift directions capturing accent-induced representation shifts. By injecting these directions into individual layers and measuring how they align accented and standard embeddings, we derive a layer-wise accent sensitivity profile, revealing that accent information concentrates in a narrow band of middle encoder layers. Leveraging this structure, we further introduce parameter-free accent steering that modifies representations during inference without updating model weights. Experiments across eight accents show consistent word error rate reductions.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Automatic Speech Recognition on VCTK (test)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-vctk-test.md>): Automatic Speech Recognition; VCTK (test); WER: 3.47.

- [Automatic Speech Recognition on L2-ARCTIC (test)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-l2-arctic-test.md>): Automatic Speech Recognition; L2-ARCTIC (test); WER: 9.39.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.05813>)

[PDF](<https://arxiv.org/pdf/2603.05813.pdf>)
