paper-with-me

Papers

AdapterFusion: Non-Destructive Task Composition for Transfer Learning

2020-05-01 · EACL 2021 2 · Jonas Pfeiffer, Aishwarya Kamath, Andreas Rücklé, Kyunghyun Cho, Iryna Gurevych

Sequential fine-tuning and multi-task learning are methods aiming to incorporate knowledge from multiple tasks; however, they suffer from catastrophic forgetting and difficulties in dataset balancing. To address these shortcomings, we propose AdapterFusion, a new two stage learning algorithm that leverages knowledge from multiple tasks. First, in the knowledge extraction stage we learn task specific parameters called adapters, that encapsulate the task-specific information. We then combine the adapters in a separate knowledge composition step. We show that by separating the two stages, i.e., knowledge extraction and knowledge composition, the classifier can effectively exploit the representations learned from multiple tasks in a non-destructive manner. We empirically evaluate AdapterFusion on 16 diverse NLU tasks, and find that it effectively combines various types of knowledge at different layers of the model. We show that our approach outperforms traditional strategies such as full fine-tuning as well as multi-task learning. Our code and adapters are available at AdapterHub.ml.

📄 PDF Abstract BibTeX arXiv:2005.00247

Code (3)

Adapter-Hub/adapter-transformers 공식 구현 pytorch
nlp-uoregon/trankit pytorch
sagjounkani/Dont-Stop-Pretraining-Use-Adapters-Instead pytorch

Tasks

Language ModellingMulti-Task LearningTransfer Learning

Similar Papers 제목 키워드 기반

Audio-AdapterFusion: A Task-ID-free Approach for Efficient and Non-Destructive Multi-task Speech Recognition

2023-10-17 · Hillary Ngai, Rohan Agrawal, Neeraj Gaur, Ronny Huang 외

Adapters are an efficient, composable alternative to full fine-tuning of pre-trained models and help scale the deployment of large ASR models to many tasks. In practice, a task ID is commonly prepended to the input durin…

speech-recognitionSpeech Recognition

Analysis of AdvFusion: Adapter-based Multilingual Learning for Code Large Language Models

2025-11-03 · Amirreza Esmaeili, Fahd Seddik, Yongyi Ji, Fatemeh Fard 외 arxiv

Programming languages can benefit from one another by utilizing a language model for software engineering tasks. Full fine-tuning and Parameter Efficient Fine-Tuning (PEFT) of Code Language Models (Code-LMs) has been exp…

Method name predictionCode TranslationCode Generation

MerA: Merging Pretrained Adapters For Few-Shot Learning

2023-08-30 · Shwai He, Run-Ze Fan, Liang Ding, Li Shen 외

Adapter tuning, which updates only a few parameters, has become a mainstream method for fine-tuning pretrained language models to downstream tasks. However, it often yields subpar results in few-shot learning. AdapterFus…

Few-Shot LearningMRPC

Pruning Adatperfusion with Lottery Ticket Hypothesis

2022-01-16 · ACL ARR January 2022 1 · Anonymous

Pre-trained language models have shown great success in multiple downstream tasks. However, they are computationally expensive to fine-tune. Thus, transfer learning with adapter modules has been introduced to alleviate t…

Transfer Learning

ScaLearn: Simple and Highly Parameter-Efficient Task Transfer by Learning to Scale

2023-10-02 · Markus Frohmann, Carolin Holtermann, Shahed Masoudian, Anne Lauscher 외

Multi-task learning (MTL) has shown considerable practical benefits, particularly when using language models (LMs). While this is commonly achieved by learning $n$ tasks under a joint optimization procedure, some methods…

Multi-Task LearningTransfer Learning