IBM / PoWER-BERT
Method to improve inference time for BERT. This is an implementation of the paper titled "PoWER-BERT: Accelerating BERT Inference via Progressive Word-vector Elimination" accepted in ICML 2020.
☆58Updated last year
Related projects ⓘ
Alternatives and complementary repositories for PoWER-BERT
- ☆47Updated 4 years ago
- Pytorch implementation of paper "Efficient Nearest Neighbor Language Models" (EMNLP 2021)☆71Updated 2 years ago
- Source code for NAACL 2021 paper "TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference"☆44Updated 2 years ago
- Official Pytorch Implementation of Length-Adaptive Transformer (ACL 2021)☆100Updated 4 years ago
- Code for EMNLP 2020 paper CoDIR☆41Updated 2 years ago
- Code for the paper "A Theoretical Analysis of the Repetition Problem in Text Generation" in AAAI 2021.☆51Updated 2 years ago
- ☆32Updated 3 years ago
- Code for the paper "BERT Loses Patience: Fast and Robust Inference with Early Exit".☆64Updated 3 years ago
- Code for the paper "Are Sixteen Heads Really Better than One?"☆169Updated 4 years ago
- Implementation of ICLR 2020 paper "Revisiting Self-Training for Neural Sequence Generation"☆47Updated 2 years ago
- [NeurIPS 2020] "The Lottery Ticket Hypothesis for Pre-trained BERT Networks", Tianlong Chen, Jonathan Frankle, Shiyu Chang, Sijia Liu, Ya…☆138Updated 2 years ago
- This repository contains the code for running the character-level Sandwich Transformers from our ACL 2020 paper on Improving Transformer …☆55Updated 3 years ago
- Source code for paper: Knowledge Inheritance for Pre-trained Language Models☆38Updated 2 years ago
- ☆40Updated 3 years ago
- ☆42Updated 4 years ago
- ☆42Updated 3 years ago
- Sequence-Level Mixed Sample Data Augmentation☆21Updated 3 years ago
- ☆19Updated 4 years ago
- DisCo Transformer for Non-autoregressive MT☆78Updated 2 years ago
- The implementation of "Neural Machine Translation without Embeddings", NAACL 2021☆33Updated 3 years ago
- ICLR2019, Multilingual Neural Machine Translation with Knowledge Distillation☆70Updated 4 years ago
- ☆30Updated 4 years ago
- ☆42Updated 3 years ago
- ENGINE: Energy-Based Inference Networks for Non-Autoregressive Machine Translation☆24Updated 4 years ago
- ☆95Updated 2 years ago
- Code for our EMNLP 2020 Paper "AIN: Fast and Accurate Sequence Labeling with Approximate Inference Network"☆18Updated 2 years ago
- ☆21Updated 3 years ago
- PyTorch implementation of NAACL 2021 paper "Multi-view Subword Regularization"☆24Updated 3 years ago
- ☆44Updated 3 years ago
- ReConsider is a re-ranking model that re-ranks the top-K (passage, answer-span) predictions of an Open-Domain QA Model like DPR (Karpukhi…☆49Updated 3 years ago