Efficient Cluster-Based k-Nearest-Neighbor Machine Translation. - Search

Open links in new tab

Copilot Answer
ACL Anthology
https://aclanthology.org
Efficient Cluster-Based k -Nearest-Neighbor Machine Translation
To make it practical, in this paper, we explore a more efficient k NN-MT and propose to use clustering to improve the retrieval efficiency. Concretely, we first propose a cluster-based Compact Network for feature reduction in a contrastive learning manner to compress context features into 90+% lower di…
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
Our proposed methods achieve better or comparable performance while reducing up to 57% inference latency against the advanced non-parametric MT model o…
arXiv.org
[2010.00710] Nearest Neighbor Machine Translation - arXiv.org
We introduce k -nearest-neighbor machine translation (k NN-MT), which predicts tokens with a nearest neighbor classifier over a large datastore of cach…
arXiv.org
Can a cluster-based compact network reduce context features?
Concretely, we first propose a cluster-based Compact Network for feature reduction in a contrastive learning manner to compress context features into 90+% lower dimensional vectors. We then suggest a cluster-based pruning solution to filter out 10%-40% redundant nodes in large datastores while retaining translation quality.
arxiv.org
Watch video
2:13
K Nearest Neighbors | Intuitive explained | Machine Learning Basics
YouTubeIntuitive Machine Learning246.6K viewsApr 26, 2020

To make it practical, in this paper, we explore a more efficient kNN-MT and propose to use clustering to improve the retrieval efficiency.
Cite as: arXiv:2204.06175 [cs.CL]
Subjects: Computation and Language (cs.CL)
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
arxiv.org/abs/2204.06175
Was this helpful?
arXiv.org
https://arxiv.org › abs
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
Apr 13, 2022 · Our proposed methods achieve better or comparable performance while reducing up to 57% inference latency against the advanced non-parametric MT model on several …
- Cite as: arXiv:2204.06175 [cs.CL]
- Subjects: Computation and Language (cs.CL)
ACL Anthology
https://aclanthology.org
[PDF]
Efcient Cluster-Based k-Nearest-Neighbor Machine Translation
Meta-k Network Prediction. Meta-k Network
Datastore distribution
X p(tijs; t<i) ^ = p (kr) pkrNN(tijs; t<i) ^
Algorithm 1 Cluster-Based Pruning Input:
Dnew fg.
4 Experiments
4.5 Results
D Case Analysis
(f ) is a two-layer feed-forward network followed by a non-linear activation function. Based on the
See more on aclanthology.org
Semantic Scholar
https://www.semanticscholar.org › paper › Efficient...
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
Apr 13, 2022 · This paper proposes a simply trainable nearest neighbor machine translation and carries out inference experiments on GPU, which demonstrates that knnMT can be integrated …
Github
https://github.com › wonderseen › PCKMT
GitHub - wonderseen/PCKMT: Source codes of ACL 2022-Efficient …
Source codes of ACL 2022-Efficient Cluster-Based k-Nearest-Neighbor Machine Translation. The implement of our proposed PCKMT is build upon the research of: adaptive kNN-MT (Xin …
ar5iv
https://ar5iv.labs.arxiv.org › abs
Efficient Cluster-Based 𝑘-Nearest-Neighbor Machine Translation
Efficient Cluster-Based k 𝑘 k-Nearest-Neighbor Machine Translation. Dexin Wang Tianjin University dexinwang@tju.edu.cn &Kai Fan * interfk@gmail.com \AND Boxing Chen …
People also ask
What is k -nearest-neighbor machine translation (k NN-Mt)?
We introduce k -nearest-neighbor machine translation (k NN-MT), which predicts tokens with a nearest neighbor classifier over a large datastore of cached examples, using representations from a neural translation model for similarity search.
[2010.00710] Nearest Neighbor Machine Translation - arXiv.org
arxiv.org
Can a cluster-based pruning solution reduce machine translation latency?
We then suggest a cluster-based pruning solution to filter out 10% 40% redundant nodes in large datastores while retaining translation quality. Our proposed methods achieve better or comparable performance while reducing up to 57% inference latency against the advanced non-parametric MT model on several machine translation benchmarks.
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
aclanthology.org
Can a cluster-based compact network reduce context features?
Concretely, we first propose a cluster-based Compact Network for feature reduction in a contrastive learning manner to compress context features into 90+% lower dimensional vectors. We then suggest a cluster-based pruning solution to filter out 10%-40% redundant nodes in large datastores while retaining translation quality.
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
arxiv.org
Is k-nearest-neighbor machine translation a nonparametric solution for domain adaptation?
k-Nearest-Neighbor Machine Translation (kNN-MT) has been recently proposed as a non-parametric solution for domain adaptation in neural machine translation (NMT).
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
aclanthology.org
Can clustering improve KNN retrieval efficiency?
In spite of this success, kNN retrieval is at the expense of high latency, in particular for large datastores. To make it practical, in this paper, we explore a more efficient kNN-MT and propose to use clustering to improve the retrieval efficiency.
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
aclanthology.org
Does the Compact network improve the retrieval for machine translation?
CKMT* gains an average improvement of 0.70 BLEU over the adaptive kNN-MT which indicates that our proposed Compact Network refines the retrieval for machine translation. The Compact Network Training with Lim-ited Data. It is unclear how much data are ade-quate at training-stage I.
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation
arxiv.org
Feedback
ResearchGate
https://www.researchgate.net › publication
Efficient Cluster-Based k-Nearest-Neighbor Machine Translation …
Jan 1, 2022 · In this paper, we propose a simple and scalable nearest neighbor machine translation framework to drastically promote the decoding and storage efficiency of $k$NN …
arXiv.org
https://arxiv.org › abs
[2010.00710] Nearest Neighbor Machine Translation - arXiv.org
Oct 1, 2020 · We introduce k -nearest-neighbor machine translation (k NN-MT), which predicts tokens with a nearest neighbor classifier over a large datastore of cached examples, using …
arXiv.org
https://arxiv.org › pdf
[PDF]
Efﬁcient Cluster-Based k-Nearest-Neighbor Machine …
compress context features into 90+% lower dimensional vectors. We then suggest a cluster-based pruning solution to filter out 10%~40% redundan. nodes in large datastores while …
Github
https://github.com › tjunlp-lab › PCKMT
Source codes of ACL 2022-Efficient Cluster-Based k …
Source codes of ACL 2022-Efficient Cluster-Based k-Nearest-Neighbor Machine Translation Resources
Some results have been removed
Pagination
- 1
- 2
- 3
- 4
- Next