Research map / Probabilistic and scientific modeling / Variational and generative sampling
Variational Recurrent Neural Machine Translation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use recurrent timestep-wise latent variables instead of a deterministic decoder | Attention-based NMT · Chinese-English MT03/04/05/06 | BLEU: 5.66 BLEU points over Moses | Helped |
| Use recurrent timestep-wise latent variables instead of a single global latent variable | VNMT · Chinese-English MT03/04/05/06 | BLEU: 1.42, 0.78, 1.0 BLEU points over ... DL4MT, VNMT and VRNMT(-TD) | Helped |
| Add temporal dependencies between latent variables (ablation) | VRNMT · Chinese-English MT03/04/05/06 | BLEU: 1.0 BLEU | Helped |
| Use recurrent timestep-wise latent variables instead of a single global latent variable | VNMT · English-German news-test 2015 | BLEU: 0.44 BLEU points | Helped |
| Add temporal dependencies between latent variables (ablation) | VRNMT · English-German news-test 2015 | BLEU: 0.59 BLEU points | Helped |
| Add recurrent latent variables to the decoder | NMT · Chinese-English translation | N-GRR 1-Gram: 19.62 -> 18.83 | Helped |
| Add recurrent latent variables to the decoder | NMT · Chinese-English word alignment | AER: 50.07 -> 48.11 | Helped |
Closest papers by meaning
- IntroVNMT: An Introspective Model for Variational Neural Machine Translation (AAAI 2020 · similarity 0.89)
- Generative Neural Machine Translation (NeurIPS 2018 · similarity 0.89)
- Regularizing Neural Machine Translation by Target-Bidirectional Agreement (AAAI 2019 · similarity 0.88)
- Variational Smoothing in Recurrent Neural Network Language Models (ICLR 2019 · similarity 0.88)
- Latent-Variable Non-Autoregressive Neural Machine Translation with Deterministic Inference Using a Delta Posterior (AAAI 2020 · similarity 0.88)
- Towards Making the Most of BERT in Neural Machine Translation (AAAI 2020 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.