IDSIA/recurrent-fwp
Official repository for the paper "Going Beyond Linear Transformers with Recurrent Fast Weight Programmers" (NeurIPS 2021)
This project offers an advanced method for building more efficient and capable AI models, particularly for tasks requiring strong pattern recognition and sequential processing. It takes in complex datasets used for training AI and produces enhanced models that can perform better on various computational tasks. AI researchers and machine learning engineers developing new deep learning architectures would use this to push the boundaries of current model performance.
No commits in the last 6 months.
Use this if you are an AI researcher or machine learning engineer looking to implement cutting-edge deep learning architectures that can outperform traditional transformer models on tasks like language modeling or reinforcement learning.
Not ideal if you are an end-user seeking an off-the-shelf application or a readily deployable solution, as this project provides foundational research code for model development rather than a finished product.
Stars
51
Forks
5
Language
Python
License
—
Category
Last pushed
Jun 11, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/IDSIA/recurrent-fwp"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
galilai-group/stable-pretraining
Reliable, minimal and scalable library for pretraining foundation and world models
CognitiveAISystems/MAPF-GPT
[AAAI-2025] This repository contains MAPF-GPT, a deep learning-based model for solving MAPF...
UKPLab/gpl
Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled...
larslorch/avici
Amortized Inference for Causal Structure Learning, NeurIPS 2022
svdrecbd/mhc-mlx
MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.