FSA-GRPO: Teaching Auditory LLMs to Use Few-shot Demonstrations

• URL: http://arxiv.org/abs/2606.02615v1
• Date: Tue, 26 May 2026 15:36:07 GMT
• Title: FSA-GRPO: Teaching Auditory LLMs to Use Few-shot Demonstrations
• Abstract summary: Few-Shot Aware GRPO is a post-training recipe that uses a specially designed reward to encourage the model to leverage few-shot demonstrations.<n>We show that training with only high-resource adult ASR data improves the model's general few-shot adaptation ability.
• Score: 33.5309824570101
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Few-shot prompting provides an effective way to adapt auditory large language models to low-resource tasks such as children's speech recognition. However, most auditory large language models are not explicitly trained to perform inference in this demonstration-conditioned format, limiting the extent to which they can benefit from few-shot prompting. To address this limitation, we introduce Few-Shot Aware GRPO (FSA-GRPO), an RL-based post-training recipe that uses a specially designed reward to encourage the model to leverage few-shot demonstrations, thereby strengthening its few-shot adaptation ability. Notably, training with only high-resource adult ASR data improves the model's general few-shot adaptation ability, yielding gains not only in children's speech recognition but also in speech translation and audio understanding. We further study data selection and auxiliary reward weighting to identify an effective training recipe. Our experiments show that when in-domain data are unavailable or cannot be used for training, FSA-GRPO is more effective than direct tuning on related out-of-domain data.

Related papers

• SICL-AT: Another way to adapt Auditory LLM to low-resource task [34.82834349882226]
  Auditory Large Language Models (LLMs) have demonstrated strong performance across a wide range of speech and audio understanding tasks.<n>They often struggle when applied to low-resource or unfamiliar tasks.<n>In-Context Learning (ICL) provides a training-free, inference-time solution.
  arXiv  Detail & Related papers  (2026-01-26T19:15:16Z)
• Omni-AVSR: Towards Unified Multimodal Speech Recognition with Large Language Models [34.15708407614003]
  Large language models (LLMs) have recently achieved impressive results in speech recognition across multiple modalities.<n>We present Omni-AVSR, a unified audio-visual LLM that combines efficient multi-granularity training with parameter-efficient adaptation.<n> Experiments on LRS2 and LRS3 show that Omni-AVSR achieves comparable or superior accuracy to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-11-10T16:03:44Z)
• Customizing Speech Recognition Model with Large Language Model Feedback [5.290365603660415]
  We propose a reinforcement learning based approach for unsupervised domain adaptation.<n>We leverage unlabeled data to enhance transcription quality, particularly the named entities affected by domain mismatch.<n>Our method achieves a 21% improvement on entity word error rate over conventional self-training methods.
  arXiv  Detail & Related papers  (2025-06-05T18:42:57Z)
• Analyzing Mitigation Strategies for Catastrophic Forgetting in End-to-End Training of Spoken Language Models [79.90523648823522]
  Multi-stage continual learning can lead to catastrophic forgetting.<n>This paper evaluates three mitigation strategies-model merging, discounting the LoRA scaling factor, and experience replay.<n>Results show that experience replay is the most effective, with further gains achieved by combining it with other methods.
  arXiv  Detail & Related papers  (2025-05-23T05:50:14Z)
• MSA-ASR: Efficient Multilingual Speaker Attribution with frozen ASR Models [59.80042864360884]
  Speaker-attributed automatic speech recognition (SA-ASR) aims to transcribe speech while assigning transcripts to the corresponding speakers accurately.<n>This paper introduces a novel approach, leveraging a frozen multilingual ASR model to incorporate speaker attribution into the transcriptions.
  arXiv  Detail & Related papers  (2024-11-27T09:01:08Z)
• SMILE: Speech Meta In-Context Learning for Low-Resource Language Automatic Speech Recognition [55.2480439325792]
  Speech Meta In-Context LEarning (SMILE) is an innovative framework that combines meta-learning with speech in-context learning (SICL)<n>We show that SMILE consistently outperforms baseline methods in training-free few-shot multilingual ASR tasks.
  arXiv  Detail & Related papers  (2024-09-16T16:04:16Z)
• Self-supervised Adaptive Pre-training of Multilingual Speech Models for Language and Dialect Identification [19.893213508284813]
  Self-supervised adaptive pre-training is proposed to adapt the pre-trained model to the target domain and languages of the downstream task. We show that SAPT improves XLSR performance on the FLEURS benchmark with substantial gains up to 40.1% for under-represented languages.
  arXiv  Detail & Related papers  (2023-12-12T14:58:08Z)
• Reduce, Reuse, Recycle: Is Perturbed Data better than Other Language augmentation for Low Resource Self-Supervised Speech Models [48.44820587495038]
  Self-supervised representation learning (SSRL) has demonstrated superior performance than supervised models for tasks including phoneme recognition. Training SSRL models poses a challenge for low-resource languages where sufficient pre-training data may not be available. We propose to use audio augmentation techniques, namely: pitch variation, noise addition, accented target language and other language speech to pre-train SSRL models in a low resource condition and evaluate phoneme recognition.
  arXiv  Detail & Related papers  (2023-09-22T10:09:09Z)
• An Exploration of Self-Supervised Pretrained Representations for End-to-End Speech Recognition [98.70304981174748]
  We focus on the general applications of pretrained speech representations, on advanced end-to-end automatic speech recognition (E2E-ASR) models. We select several pretrained speech representations and present the experimental results on various open-source and publicly available corpora for E2E-ASR.
  arXiv  Detail & Related papers  (2021-10-09T15:06:09Z)
• Don't Stop Pretraining: Adapt Language Models to Domains and Tasks [81.99843216550306]
  We present a study across four domains (biomedical and computer science publications, news, and reviews) and eight classification tasks. A second phase of pretraining in-domain (domain-adaptive pretraining) leads to performance gains. Adapting to the task's unlabeled data (task-adaptive pretraining) improves performance even after domain-adaptive pretraining.
  arXiv  Detail & Related papers  (2020-04-23T04:21:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.