# Neural Machine Translation with Gumbel-Greedy Decoding

arXiv: https://arxiv.org/abs/1706.07518 · HTML: https://mlsanity.com/paper/1706.07518/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2017.

## Built on by (most cited first, 5 of 5)

- [Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution](https://mlsanity.com/paper/2106.06361/) (2021)
- [A Review of the Gumbel-max Trick and its Extensions for Discrete Stochasticity in Machine Learning](https://mlsanity.com/paper/2110.01515/) (2021)
- [Soft-Di[M]O: Improving One-Step Discrete Image Generation with Soft Embeddings](https://mlsanity.com/paper/2509.22925/) (2025)
- [Sequential Visual and Semantic Consistency for Semi-supervised Text Recognition](https://mlsanity.com/paper/2402.15806/) (2024)
- [SemanticST: Spatially Informed Semantic Graph Learning for Clustering, Integration, and Scalable Analysis of Spatial Transcriptomics](https://mlsanity.com/paper/2506.11491/) (2025)

## Abstract (arXiv)

Previous neural machine translation models used some heuristic search algorithms (e.g., beam search) in order to avoid solving the maximum a posteriori problem over translation sentences at test time. In this paper, we propose the Gumbel-Greedy Decoding which trains a generative network to predict translation under a trained model. We solve such a problem using the Gumbel-Softmax reparameterization, which makes our generative network differentiable and trainable through standard stochastic gradient methods. We empirically demonstrate that our proposed model is effective for generating sequences of discrete words.