twistedcubic / attention-rank-collapseLinks
[ICML 2021 Oral] We show pure attention suffers rank collapse, and how different mechanisms combat it.
☆166Updated 4 years ago
Alternatives and similar repositories for attention-rank-collapse
Users that are interested in attention-rank-collapse are comparing it to the libraries listed below
Sorting:
- Official PyTorch Implementation of Long-Short Transformer (NeurIPS 2021).☆225Updated 3 years ago
- Sinkhorn Transformer - Practical implementation of Sparse Sinkhorn Attention☆267Updated 4 years ago
- Official code repository of the paper Linear Transformers Are Secretly Fast Weight Programmers.☆105Updated 4 years ago
- [ICML 2020] code for "PowerNorm: Rethinking Batch Normalization in Transformers" https://arxiv.org/abs/2003.07845☆120Updated 4 years ago
- Code for Multi-Head Attention: Collaborate Instead of Concatenate☆151Updated 2 years ago
- [ICML 2020] code for the flooding regularizer proposed in "Do We Need Zero Training Loss After Achieving Zero Training Error?"☆92Updated 2 years ago
- [ICLR 2022] Official implementation of cosformer-attention in cosFormer: Rethinking Softmax in Attention☆196Updated 2 years ago
- Awesome Contrastive Learning for CV & NLP☆163Updated 4 years ago
- Implementation of Long-Short Transformer, combining local and global inductive biases for attention over long sequences, in Pytorch☆119Updated 4 years ago
- Fully featured implementation of Routing Transformer☆297Updated 3 years ago
- Learning to Encode Position for Transformer with Continuous Dynamical Model☆60Updated 5 years ago
- Implementing SYNTHESIZER: Rethinking Self-Attention in Transformer Models using Pytorch☆70Updated 5 years ago
- ☆81Updated last year
- Crawl & visualize ICLR papers and reviews.☆107Updated 3 years ago
- Implementation of Memformer, a Memory-augmented Transformer, in Pytorch☆119Updated 4 years ago
- ☆84Updated 4 years ago
- NeurIPS 2020, Debiased Contrastive Learning☆282Updated 2 years ago
- ☆36Updated 4 years ago
- Loss and accuracy go opposite ways...right?☆95Updated 5 years ago
- Code for the ICML'20 paper "Improving Transformer Optimization Through Better Initialization"☆89Updated 4 years ago
- Apollo: An Adaptive Parameter-wise Diagonal Quasi-Newton Method for Nonconvex Stochastic Optimization☆182Updated 3 years ago
- Understanding Training Dynamics of Deep ReLU Networks☆295Updated 3 weeks ago
- Open source code for paper "Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere" ICML 2…☆448Updated 3 years ago
- Pytorch implementation of Compressive Transformers, from Deepmind☆163Updated 3 years ago
- [EMNLP'19] Summary for Transformer Understanding☆53Updated 5 years ago
- Code to reproduce the results for Compositional Attention☆60Updated 2 years ago
- A PyTorch implementation of the paper - "Synthesizer: Rethinking Self-Attention in Transformer Models"☆73Updated 2 years ago
- A supplementary code for Editable Neural Networks, an ICLR 2020 submission.☆46Updated 5 years ago
- code for paper "Graph Structure of Neural Networks"☆155Updated 3 years ago
- Implementation of Sparsemax activation in Pytorch☆162Updated 5 years ago