This library provides functionality for rapidly sharing and retrieving word embeddings over the internet. (EMNLP 2017).
-
Updated
Dec 3, 2018 - Python
#
shared-embedding
Here are 4 public repositories matching this topic...
The Transformer model implemented from scratch using PyTorch. The model uses weight sharing between the embedding layers and the pre-softmax linear layer. Training on the Multi30k machine translation task is shown.
deep-learning machine-translation pytorch transformer shared-embedding multi-head-attention multi30k
-
Updated
Jul 23, 2023 - Python
Pytorch implementation of "Multi-domain translation between single-cell imaging and sequencing data using autoencoders" (https://www.nature.com/articles/s41467-020-20249-2) with custom models.
multi-domain single-cell multi-modal single-cell-rna-seq shared-embedding multi-view-learning single-cell-omics multi-view data-alignment multi-modal-learning multi-domain-adaptation
-
Updated
Oct 13, 2021 - Python
DNA-language hybrid model reference implementation (Carbon-X style): one decoder with a mixed tokenizer (6-mer DNA + text, shared embedding) for mutation pathogenicity, cross-modal interpretation and constrained DNA generation. Closed-form oracle + verifiable reward + monotone benchmark. Pure Python, zero deps.
python zero-dependency cross-modal shared-embedding rlhf dna-language-model verifiable-reward genomic-foundation-model mixed-tokenizer 6-mer mutation-pathogenicity
-
Updated
Nov 4, 2026 - Python
Add this topic to your repo
To associate your repository with the shared-embedding topic, visit your repo's landing page and select "manage topics."