Search Results - "Passban, Peyman"

Improving Phrase-Based SMT Using Cross-Granularity Embedding Similarity by Passban, Peyman, Hokamp, Chris, Way, Andy, Liu, Qun

“…The phrase-based statistical machine translation (PBSMT) model can be viewed as a log-linear combination of translation and language model features. Such a…”
Get full text

What is Lost in Knowledge Distillation? by Mohanty, Manas, Roosta, Tanya, Passban, Peyman

“…Deep neural networks (DNNs) have improved NLP tasks significantly, but training and maintaining such networks could be costly. Model compression techniques,…”
Get full text

Dynamic Position Encoding for Transformers by Zheng, Joyce, Rezagholizadeh, Mehdi, Passban, Peyman

“…Recurrent models have been dominating the field of neural machine translation (NMT) for the past few years. Transformers \citep{vaswani2017attention}, have…”
Get full text

Communication-Efficient Federated Learning for Neural Machine Translation by Roosta, Tanya, Passban, Peyman, Chadha, Ankit

“…Training neural machine translation (NMT) models in federated learning (FL) settings could be inefficient both computationally and communication-wise, due to…”
Get full text

Training Mixed-Domain Translation Models via Federated Learning by Passban, Peyman, Roosta, Tanya, Gupta, Rahul, Chadha, Ankit, Chung, Clement

“…Training mixed-domain translation models is a complex task that demands tailored architectures and costly data preparation techniques. In this work, we…”
Get full text

Not Far Away, Not So Close: Sample Efficient Nearest Neighbour Data Augmentation via MiniMax by Kamalloo, Ehsan, Rezagholizadeh, Mehdi, Passban, Peyman, Ghodsi, Ali

“…In Natural Language Processing (NLP), finding data augmentation techniques that can produce high-quality human-interpretable examples has always been…”
Get full text

Revisiting Robust Neural Machine Translation: A Transformer Case Study by Passban, Peyman, Saladi, Puneeth S. M, Liu, Qun

“…Transformers (Vaswani et al., 2017) have brought a remarkable improvement in the performance of neural machine translation (NMT) systems but they could be…”
Get full text

ALP-KD: Attention-Based Layer Projection for Knowledge Distillation by Passban, Peyman, Wu, Yimeng, Rezagholizadeh, Mehdi, Liu, Qun

“…Knowledge distillation is considered as a training and compression strategy in which two neural networks, namely a teacher and a student, are coupled together…”
Get full text

Why Skip If You Can Combine: A Simple Knowledge Distillation Technique for Intermediate Layers by Wu, Yimeng, Passban, Peyman, Rezagholizade, Mehdi, Liu, Qun

“…With the growth of computing power neural machine translation (NMT) models also grow accordingly and become better. However, they also become harder to deploy…”
Get full text

Improving Character-based Decoding Using Target-Side Morphological Information for Neural Machine Translation by Passban, Peyman, Liu, Qun, Way, Andy

“…Recently, neural machine translation (NMT) has emerged as a powerful alternative to conventional statistical approaches. However, its performance drops…”
Get full text

From Fully Trained to Fully Random Embeddings: Improving Neural Machine Translation with Compact Word Embedding Tables by Kumar, Krtin, Passban, Peyman, Rezagholizadeh, Mehdi, Lau, Yiu Sing, Liu, Qun

“…Embedding matrices are key components in neural natural language processing (NLP) models that are responsible to provide numerical representations of input…”
Get full text

Robust Embeddings Via Distributions by Selby, Kira A, Wang, Yinong, Wang, Ruizhe, Passban, Peyman, Rashid, Ahmad, Rezagholizadeh, Mehdi, Poupart, Pascal

“…Despite recent monumental advances in the field, many Natural Language Processing (NLP) models still struggle to perform adequately on noisy domains. We…”
Get full text

POSABL: ABL-based grammar induction on POS tags by Passban, Peyman, Shokrollahi-Far, Mahmoud

“…Grammar Induction (GI) is the problem of extracting hidden regularities and syntactic patterns in languages. Not only the manner of extraction is intricate but…”
Get full text

Investigating Backtranslation in Neural Machine Translation by Poncelas, Alberto, Shterionov, Dimitar, Way, Andy, Wenniger, Gideon Maillette de Buy, Passban, Peyman

“…A prerequisite for training corpus-based machine translation (MT) systems -- either Statistical MT (SMT) or Neural MT (NMT) -- is the availability of…”
Get full text

A reliable linguistic filter for Farsi term extraction by Behin-Faraz, Hossein, Passban, Peyman, Shokrollahi-Far, Mahmoud

“…Ontology learning has four main successive mining phases: term, concept, relation and axiom extraction. Since term extraction (TE) is the foundation of the…”
Get full text

Developing a shuffle grammar for parsing Arabic verbs by Passban, P., Shokrollahi-Far, M.

“…One of the most important members of natural languages is Arabic. This language spans 14 centuries and 22 countries and has a great traditional and historical…”
Get full text