paper-with-me

Papers

Investigating Efficiently Extending Transformers for Long Input Summarization

2022-08-08 · Jason Phang, Yao Zhao, Peter J. Liu

While large pretrained Transformer models have proven highly capable at tackling natural language tasks, handling long sequence inputs continues to be a significant challenge. One such task is long input summarization, where inputs are longer than the maximum input context of most pretrained models. Through an extensive set of experiments, we investigate what model architectural changes and pretraining paradigms can most efficiently adapt a pretrained Transformer for long input summarization. We find that a staggered, block-local Transformer with global encoder tokens strikes a good balance of performance and efficiency, and that an additional pretraining phase on long sequences meaningfully improves downstream summarization performance. Based on our findings, we introduce PEGASUS-X, an extension of the PEGASUS model with additional long input pretraining to handle inputs of up to 16K tokens. PEGASUS-X achieves strong performance on long input summarization tasks comparable with much larger models while adding few additional parameters and not requiring model parallelism to train.

📄 PDF Abstract BibTeX arXiv:2208.04347

Code (2)

google-research/pegasus 공식 구현 tf
MindCode-4/code-2/tree/main/pegasus_x mindspore

Tasks

16kLong-range modelingText Summarization

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
PEGASUS PEGASUS proposes a transformer-based model for abstractive summarization. It uses a special self-supervised pre-training objective called gap-sentences generation (GSG) that's…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
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…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Functional Interpolation for Relative Positions Improves Long Context Transformers

2023-10-06 · Shanda Li, Chong You, Guru Guruganesh, Joshua Ainslie 외

Preventing the performance decay of Transformers on inputs longer than those used for training has been an important challenge in extending the context length of these models. Though the Transformer architecture has fund…

Language ModelingLanguage ModellingPosition

Unlimiformer: Long-Range Transformers with Unlimited Length Input

2023-05-02 · NeurIPS 2023 11 · Amanda Bertsch, Uri Alon, Graham Neubig, Matthew R. Gormley

Since the proposal of transformers, these models have been limited to bounded input lengths, because of their need to attend to every token in the input. In this work, we propose Unlimiformer: a general approach that wra…

Book summarizationCPUDecoderGPU

Investigating Recurrent Transformers with Dynamic Halt

2024-02-01 · Jishnu Ray Chowdhury, Cornelia Caragea

In this paper, we comprehensively study the inductive biases of two major approaches to augmenting Transformers with a recurrent mechanism: (1) the approach of incorporating a depth-wise recurrence similar to Universal T…

DiagnosticLanguage ModelingLanguage ModellingListOps

Curved Representation Space of Vision Transformers

2022-10-11 · Juyeop Kim, Junha Park, Songkuk Kim, Jong-Seok Lee

Neural networks with self-attention (a.k.a. Transformers) like ViT and Swin have emerged as a better alternative to traditional convolutional neural networks (CNNs). However, our understanding of how the new architecture…

SWAT: Scalable and Efficient Window Attention-based Transformers Acceleration on FPGAs

2024-05-27 · Zhenyu Bai, Pranav Dangi, Huize Li, Tulika Mitra

Efficiently supporting long context length is crucial for Transformer models. The quadratic complexity of the self-attention computation plagues traditional Transformers. Sliding window-based static sparse attention miti…

GPU