paper-with-me

홈 › Papers

Robustly Optimized and Distilled Training for Natural Language Understanding

2021-03-16 · Haytham ElFadeel, Stan Peshterliev

In this paper, we explore multi-task learning (MTL) as a second pretraining step to learn enhanced universal language representation for transformer language models. We use the MTL enhanced representation across several natural language understanding tasks to improve performance and generalization. Moreover, we incorporate knowledge distillation (KD) in MTL to further boost performance and devise a KD variant that learns effectively from multiple teachers. By combining MTL and KD, we propose Robustly Optimized and Distilled (ROaD) modeling framework. We use ROaD together with the ELECTRA model to obtain state-of-the-art results for machine reading comprehension and natural language inference.

📄 PDF Abstract BibTeX arXiv:2103.08809

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationMachine Reading ComprehensionMulti-Task LearningNatural Language InferenceNatural Language UnderstandingReading Comprehension

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Weight Decay 설명 없음
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Residual Connection 설명 없음
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

Transformer-based approaches to Sentiment Detection

2023-03-13 · Olumide Ebenezer Ojo, Hoang Thang Ta, Alexander Gelbukh, Hiram Calvo 외

The use of transfer learning methods is largely responsible for the present breakthrough in Natural Learning Processing (NLP) tasks across multiple domains. In order to solve the problem of sentiment detection, we examin…

text-classificationText ClassificationTransfer Learning

Leveraging Large Language Models for Enhanced NLP Task Performance through Knowledge Distillation and Optimized Training Strategies

2024-02-14 · Yining Huang, Keke Tang, Meilian Chen

Emerging Large Language Models (LLMs) like GPT-4 have revolutionized Natural Language Processing (NLP), showing potential in traditional tasks such as Named Entity Recognition (NER). Our study explores a three-phase trai…

Knowledge Distillationnamed-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)+1

StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding

2019-08-13 · ICLR 2020 1 · Wei Wang, Bin Bi, Ming Yan, Chen Wu 외

Recently, the pre-trained language model, BERT (and its robustly optimized version RoBERTa), has attracted a lot of attention in natural language understanding (NLU), and achieved state-of-the-art accuracy in various NLU…

Language ModelingLanguage ModellingLinguistic AcceptabilityNatural Language Inference+7

Llamba: Scaling Distilled Recurrent Models for Efficient Language Processing

2025-02-20 · Aviv Bick, Tobias Katsch, Nimit Sohoni, Arjun Desai 외

We introduce Llamba, a family of efficient recurrent language models distilled from Llama-3.x into the Mamba architecture. The series includes Llamba-1B, Llamba-3B, and Llamba-8B, which achieve higher inference throughpu…

Mamba

Alexa Teacher Model: Pretraining and Distilling Multi-Billion-Parameter Encoders for Natural Language Understanding Systems

2022-06-15 · Jack FitzGerald, Shankar Ananthakrishnan, Konstantine Arkoudas, Davide Bernardi 외

We present results from a large-scale experiment on pretraining encoders with non-embedding parameter counts ranging from 700M to 9.3B, their subsequent distillation into smaller models ranging from 17M-170M parameters, …

Cross-Lingual Natural Language Inferenceintent-classificationIntent ClassificationNatural Language Inference+4