[ICML 2021 Oral] We show pure attention suffers rank collapse, and how different mechanisms combat it.
☆172Mar 8, 2021Updated 5 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. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Experts☆16Feb 26, 2024Updated 2 years ago
- ☆12Sep 26, 2019Updated 6 years ago
- We investigated corruption robustness across different architectures including Convolutional Neural Networks, Vision Transformers, and th…☆16Oct 28, 2021Updated 4 years ago
- Locally Enhanced Self-Attention: Rethinking Self-Attention as Local and Context Terms☆20Nov 29, 2021Updated 4 years ago
- Code for the TCS paper "On the performance of learned data structures" and the ICML paper "Why are learned indexes so effective?"☆21May 9, 2021Updated 5 years ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- Code Release for the 2023 NeurIPS Paper How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained langua…☆17Dec 6, 2024Updated last year
- ☆391Oct 18, 2023Updated 2 years ago
- [NeurIPS'20] Code for the Paper Compositional Visual Generation and Inference with Energy Based Models☆47Mar 24, 2023Updated 3 years ago
- A study of performance of optimal transport.☆10Jul 4, 2020Updated 5 years ago
- Implementation of the GBST block from the Charformer paper, in Pytorch☆118Jul 15, 2021Updated 4 years ago
- A simple Transformer where the softmax has been replaced with normalization☆20Sep 11, 2020Updated 5 years ago
- Variational Walkback, NIPS'17☆28Oct 18, 2017Updated 8 years ago
- ☆42Mar 23, 2023Updated 3 years ago
- ☆26Nov 23, 2023Updated 2 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- ☆27Feb 26, 2026Updated 3 months ago
- ☆20Feb 26, 2021Updated 5 years ago
- ☆122Feb 11, 2025Updated last year
- Official codebase for Pretrained Transformers as Universal Computation Engines.☆245Jan 14, 2022Updated 4 years ago
- Implementation of N-Grammer, augmenting Transformers with latent n-grams, in Pytorch☆76Dec 4, 2022Updated 3 years ago
- ☆19Jun 21, 2025Updated 11 months ago
- RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder☆210Mar 18, 2021Updated 5 years ago
- The source codes for D2AGE model. Distance-aware DAG Embedding for Proximity Search on Heterogeneous Graphs.☆12Feb 20, 2018Updated 8 years ago
- ☆45Mar 3, 2023Updated 3 years ago
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- Implementation of Fast Transformer in Pytorch☆176Aug 26, 2021Updated 4 years ago
- [CVPR 2021] Involution: Inverting the Inherence of Convolution for Visual Recognition, a brand new neural operator☆1,312Jul 16, 2021Updated 4 years ago
- Implementation of H-Transformer-1D, Hierarchical Attention for Sequence Learning☆166Feb 12, 2024Updated 2 years ago
- End-to-end training of sparse deep neural networks with little-to-no performance loss.☆337Jan 26, 2023Updated 3 years ago
- Fine-Tuning Pre-trained Transformers into Decaying Fast Weights☆19Oct 9, 2022Updated 3 years ago
- ☆46Apr 13, 2022Updated 4 years ago
- computation of convolutional kernels (CKN and NTK) in C++☆14Dec 13, 2022Updated 3 years ago
- Disentangled Non-Local Neural Networks☆84Dec 7, 2020Updated 5 years ago
- ☆20Mar 22, 2024Updated 2 years ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Transformer based on a variant of attention that is linear complexity in respect to sequence length☆838May 5, 2024Updated 2 years ago
- This is the official GitHub for paper: On the Versatile Uses of Partial Distance Correlation in Deep Learning, in ECCV 2022☆176Jun 4, 2023Updated 3 years ago
- Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch☆98Feb 19, 2021Updated 5 years ago
- Official repository for the "Big Transfer (BiT): General Visual Representation Learning" paper.☆1,541Jul 30, 2024Updated last year
- [NeurIPS‘2021] "TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up", Yifan Jiang, Shiyu Chang, Zhangyang Wang☆1,692Nov 3, 2022Updated 3 years ago
- Code for 'Contrastive Multi-Document Question Generation'☆11Oct 16, 2022Updated 3 years ago
- Official DeiT repository☆4,348Mar 15, 2024Updated 2 years ago