paper-with-me

Papers

Hierarchical Recurrent Adapters for Efficient Multi-Task Adaptation of Large Speech Models

2024-03-25 · Tsendsuren Munkhdalai, Youzheng Chen, Khe Chai Sim, Fadi Biadsy, Tara Sainath, Pedro Moreno Mengibar

Parameter efficient adaptation methods have become a key mechanism to train large pre-trained models for downstream tasks. However, their per-task parameter overhead is considered still high when the number of downstream tasks to adapt for is large. We introduce an adapter module that has a better efficiency in large scale multi-task adaptation scenario. Our adapter is hierarchical in terms of how the adapter parameters are allocated. The adapter consists of a single shared controller network and multiple task-level adapter heads to reduce the per-task parameter overhead without performance regression on downstream tasks. The adapter is also recurrent so the entire adapter parameters are reused across different layers of the pre-trained model. Our Hierarchical Recurrent Adapter (HRA) outperforms the previous adapter-based approaches as well as full model fine-tuning baseline in both single and multi-task adaptation settings when evaluated on automatic speech recognition tasks.

📄 PDF Abstract BibTeX arXiv:2403.19709

Code (0)

등록된 구현이 없습니다.

Tasks

Automatic Speech Recognitionspeech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

Adapter 설명 없음

Similar Papers 제목 키워드 기반

HierAdaptMR: Cross-Center Cardiac MRI Reconstruction with Hierarchical Feature Adapters

2025-08-18 · Ruru Xu, Ilkay Oksuz arxiv

Deep learning-based cardiac MRI reconstruction faces significant domain shift challenges when deployed across multiple clinical centers with heterogeneous scanner configurations and imaging protocols. We propose HierAdap…

MRI Reconstruction

HAM: Hierarchical Adapter Merging for Scalable Continual Learning

2025-09-16 · Eric Nuertey Coleman, Luigi Quarantiello, Samrat Mukherjee, Julio Hurtado 외 arxiv

Continual learning is an essential capability of human cognition, yet it poses significant challenges for current deep learning models. The primary issue is that new knowledge can interfere with previously learned inform…

parameter-efficient fine-tuningContinual LearningTransfer Learning

Multi-Head Adapter Routing for Cross-Task Generalization

2022-11-07 · NeurIPS 2023 11 · Lucas Caccia, Edoardo Ponti, Zhan Su, Matheus Pereira 외

Parameter-efficient fine-tuning (PEFT) for cross-task generalization consists in pre-training adapters on a multi-task training set before few-shot adaptation to test tasks. Polytropon [Ponti et al., 2023] ($\texttt{Poly…

parameter-efficient fine-tuning

Learning Options for Compositional Motor Control with Adapter Banks

2026-09-15 · Sreejan Kumar, Marcelo Mattar, Lea Duncker arxiv

Learning flexible motor primitives is a hallmark of skilled motor control. Recent neuroscience theory proposes that motor primitives may be implemented as low-rank perturbations of a shared recurrent network, but leaves …

Rethinking Adapter Placement: A Dominant Adaptation Module Perspective

2026-05-07 · Suoxin Zhang, Run He, Di Fang, Xiang Tan 외 arxiv

Low-rank adaptation (LoRA) is a widely used parameter-efficient fine-tuning method that places trainable low-rank adapters into frozen pre-trained models. Recent studies show that using fewer LoRA adapters may still main…

parameter-efficient fine-tuningMathematical ReasoningInstruction FollowingCode Generation