paper-with-me

홈 › Papers

DQ-BART: Efficient Sequence-to-Sequence Model via Joint Distillation and Quantization

2022-03-21 · ACL 2022 5 · Zheng Li, Zijian Wang, Ming Tan, Ramesh Nallapati, Parminder Bhatia, Andrew Arnold, Bing Xiang, Dan Roth

Large-scale pre-trained sequence-to-sequence models like BART and T5 achieve state-of-the-art performance on many generative NLP tasks. However, such models pose a great challenge in resource-constrained scenarios owing to their large memory requirements and high latency. To alleviate this issue, we propose to jointly distill and quantize the model, where knowledge is transferred from the full-precision teacher model to the quantized and distilled low-precision student model. Empirical analyses show that, despite the challenging nature of generative tasks, we were able to achieve a 16.5x model footprint compression ratio with little performance drop relative to the full-precision counterparts on multiple summarization and QA datasets. We further pushed the limit of compression ratio to 27.7x and presented the performance-efficiency trade-off for generative tasks using pre-trained models. To the best of our knowledge, this is the first work aiming to effectively distill and quantize sequence-to-sequence pre-trained models for language generation tasks.

📄 PDF Abstract BibTeX arXiv:2203.11239

Code (2)

amazon-research/dq-bart 공식 구현 pytorch
amazon-science/dq-bart pytorch

Tasks

Knowledge DistillationModel CompressionQuantizationText Generation

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
SentencePiece 설명 없음
Adafactor Adafactor is a stochastic optimization method based on Adam that reduces memory usage while retaining the empirical benefits of…
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…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Modular Transformers: Compressing Transformers into Modularized Layers for Flexible Efficient Inference

2023-06-04 · Wangchunshu Zhou, Ronan Le Bras, Yejin Choi

Pre-trained Transformer models like T5 and BART have advanced the state of the art on a wide range of text generation tasks. Compressing these models into smaller ones has become critically important for practical use. C…

DecoderKnowledge DistillationModel CompressionNeural Network Compression+2

BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese

2021-09-20 · Nguyen Luong Tran, Duong Minh Le, Dat Quoc Nguyen

We present BARTpho with two versions, BARTpho-syllable and BARTpho-word, which are the first public large-scale monolingual sequence-to-sequence models pre-trained for Vietnamese. BARTpho uses the "large" architecture an…

Abstractive Text SummarizationDenoisingPunctuation RestorationText Summarization

Enhancing Low-Resource NMT with a Multilingual Encoder and Knowledge Distillation: A Case Study

2024-07-09 · Aniruddha Roy, Pretam Ray, Ayush Maheshwari, Sudeshna Sarkar 외

Neural Machine Translation (NMT) remains a formidable challenge, especially when dealing with low-resource languages. Pre-trained sequence-to-sequence (seq2seq) multi-lingual models, such as mBART-50, have demonstrated i…

Knowledge DistillationLanguage ModelingLanguage ModellingLow Resource NMT+3

AraBART: a Pretrained Arabic Sequence-to-Sequence Model for Abstractive Summarization

2022-03-21 · Moussa Kamal Eddine, Nadi Tomeh, Nizar Habash, Joseph Le Roux 외

Like most natural language understanding and generation tasks, state-of-the-art models for summarization are transformer-based sequence-to-sequence architectures that are pretrained on large corpora. While most existing …

Abstractive Text SummarizationDecoderNatural Language Understanding

AraBART: a Pretrained Arabic Sequence-to-Sequence Model for Abstractive Summarization

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

Like most natural language understanding and generation tasks, state-of-the-art models for summarization are transformer-based sequence-to-sequence architectures that are pretrained on large corpora. While most existing …

Abstractive Text SummarizationDecoderNatural Language Understanding