Young Jin Kim

Principal Researcher, Microsoft GenAI

出版物

表示方法:

- Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation
  
  Liliang Ren, Congcong Chen, Haoran Xu, Young Jin Kim, Adam Atkinson, Zheng Zhan, Jiankai Sun, Baolin Peng, Liyuan Liu, Shuohang Wang, Hao Cheng, Jianfeng Gao, Weizhu Chen, Yelong Shen
  
  NeurIPS 2025 | July 2025
  
  DOI
- SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation
  
  Zichong Li, Chen Liang, Zixuan Zhang, Ilgee Hong, Young Jin Kim, Weizhu Chen, Tuo Zhao
  
  COLM 2025 | July 2025
  
  Publication
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  January 2024
  
  Preprint
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  ICML 2024 | January 2024
  
  DOI Publication Publication
- FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness
  
  Young Jin Kim, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Publication
- Task-Based MoE for Multitask Multilingual Machine Translation
  
  Hai Pham, Young Jin Kim, Subhabrata Mukherjee, David P Woodruff, Barnabas Poczos, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- PEMA: Plug-in External Memory Adaptation for Language Models
  
  HyunJin Kim, Young Jin Kim, JinYeong Bak
  
  November 2023
  
  Preprint
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
  
  Haoran Xu, Young Jin Kim, Amr Sharaf, Hany Hassan Awadalla
  
  Accepted to ICLR 2024 | September 2023
  
  Preprint
- AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks Lakshmanan, V.S., Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL 2023 | June 2023
  
  Publication
- AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL | May 2023
  
  Publication プロジェクト
- How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation
  
  Amr Hendy, Mohamed Abdelrehim, Amr Sharaf, Vikas Raunak, Mohamed Gabr, Hitokazu Matsushita, Young Jin Kim, Mohamed Afify, Hany Hassan Awadalla
  
  February 2023
  
  Preprint
- Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2022
  
  Preprint
- Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU
  
  Hossam Amer, Young Jin Kim, Mohamed Afify, Hitokazu Matsushita, Hany Hassan Awadalla
  
  AMTA | August 2022
  
  Publication プロジェクト
- Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers
  
  Rui Liu, Young Jin Kim, Alexandre Muzio, Hany Hassan Awadalla
  
  ICML 2022 | July 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan Awadalla, Ruofei Zhang, Tuo Zhao, Jianfeng Gao
  
  ICLR 2022 | April 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts.
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan, Ruofei Zhang, Tuo Zhao, Jianfeng Gao, Hany Hassan Awadalla
  
  October 2021
  
  Publication Publication
- Scalable and Efficient MoE Training for Multitask Multilingual Models
  
  Young Jin Kim, Ammar Ahmad Awan, Alexandre Muzio, Felipe Cruz Salinas, Liyang Lu, Amr Hendy, Samyam Rajbhandari, Yuxiong He, Hany Hassan Awadalla
  
  September 2021
  
  Publication Github プロジェクトプロジェクト
- FastFormers: Highly Efficient Transformer Models for Natural Language Understanding
  
  Young Jin Kim, Hany Hassan Awadalla
  
  SustaiNLP - EMNLP | November 2020
  
  Publication プロジェクト
- Time- and space-parallel simulation of air traffic networks
  
  Young Jin Kim, Dimitri Mavris, Richard Fujimoto
  
  SIMULATION (SAGE) | December 2019, 巻95(12): pp. 1213-1228
  
  Publication
- From Research to Production and Back: Ludicrously Fast Neural Machine Translation
  
  Young Jin Kim, Marcin Junczys-Dowmunt, Hany Hassan Awadalla, Alham Fikri Aji, Kenneth Heafield, Roman Grundkiewicz, Nikolay Bogoychev
  
  WNGT - EMNLP | November 2019
  
  Publication

- Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation
  
  Liliang Ren, Congcong Chen, Haoran Xu, Young Jin Kim, Adam Atkinson, Zheng Zhan, Jiankai Sun, Baolin Peng, Liyuan Liu, Shuohang Wang, Hao Cheng, Jianfeng Gao, Weizhu Chen, Yelong Shen
  
  NeurIPS 2025 | July 2025
  
  DOI
- SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation
  
  Zichong Li, Chen Liang, Zixuan Zhang, Ilgee Hong, Young Jin Kim, Weizhu Chen, Tuo Zhao
  
  COLM 2025 | July 2025
  
  Publication
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  January 2024
  
  Preprint
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  ICML 2024 | January 2024
  
  DOI Publication Publication
- Task-Based MoE for Multitask Multilingual Machine Translation
  
  Hai Pham, Young Jin Kim, Subhabrata Mukherjee, David P Woodruff, Barnabas Poczos, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness
  
  Young Jin Kim, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Publication
- PEMA: Plug-in External Memory Adaptation for Language Models
  
  HyunJin Kim, Young Jin Kim, JinYeong Bak
  
  November 2023
  
  Preprint
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
  
  Haoran Xu, Young Jin Kim, Amr Sharaf, Hany Hassan Awadalla
  
  Accepted to ICLR 2024 | September 2023
  
  Preprint
- AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks Lakshmanan, V.S., Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL 2023 | June 2023
  
  Publication
- AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL | May 2023
  
  Publication プロジェクト
- How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation
  
  Amr Hendy, Mohamed Abdelrehim, Amr Sharaf, Vikas Raunak, Mohamed Gabr, Hitokazu Matsushita, Young Jin Kim, Mohamed Afify, Hany Hassan Awadalla
  
  February 2023
  
  Preprint
- Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2022
  
  Preprint
- Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers
  
  Rui Liu, Young Jin Kim, Alexandre Muzio, Hany Hassan Awadalla
  
  ICML 2022 | July 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan Awadalla, Ruofei Zhang, Tuo Zhao, Jianfeng Gao
  
  ICLR 2022 | April 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts.
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan, Ruofei Zhang, Tuo Zhao, Jianfeng Gao, Hany Hassan Awadalla
  
  October 2021
  
  Publication Publication
- Scalable and Efficient MoE Training for Multitask Multilingual Models
  
  Young Jin Kim, Ammar Ahmad Awan, Alexandre Muzio, Felipe Cruz Salinas, Liyang Lu, Amr Hendy, Samyam Rajbhandari, Yuxiong He, Hany Hassan Awadalla
  
  September 2021
  
  Publication Github プロジェクトプロジェクト
- FastFormers: Highly Efficient Transformer Models for Natural Language Understanding
  
  Young Jin Kim, Hany Hassan Awadalla
  
  SustaiNLP - EMNLP | November 2020
  
  Publication プロジェクト
- From Research to Production and Back: Ludicrously Fast Neural Machine Translation
  
  Young Jin Kim, Marcin Junczys-Dowmunt, Hany Hassan Awadalla, Alham Fikri Aji, Kenneth Heafield, Roman Grundkiewicz, Nikolay Bogoychev
  
  WNGT - EMNLP | November 2019
  
  Publication
- AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL | May 2023
  
  Publication プロジェクト
- Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU
  
  Hossam Amer, Young Jin Kim, Mohamed Afify, Hitokazu Matsushita, Hany Hassan Awadalla
  
  AMTA | August 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan Awadalla, Ruofei Zhang, Tuo Zhao, Jianfeng Gao
  
  ICLR 2022 | April 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts.
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan, Ruofei Zhang, Tuo Zhao, Jianfeng Gao, Hany Hassan Awadalla
  
  October 2021
  
  Publication Publication
- Scalable and Efficient MoE Training for Multitask Multilingual Models
  
  Young Jin Kim, Ammar Ahmad Awan, Alexandre Muzio, Felipe Cruz Salinas, Liyang Lu, Amr Hendy, Samyam Rajbhandari, Yuxiong He, Hany Hassan Awadalla
  
  September 2021
  
  Publication Github プロジェクトプロジェクト
- FastFormers: Highly Efficient Transformer Models for Natural Language Understanding
  
  Young Jin Kim, Hany Hassan Awadalla
  
  SustaiNLP - EMNLP | November 2020
  
  Publication プロジェクト
- From Research to Production and Back: Ludicrously Fast Neural Machine Translation
  
  Young Jin Kim, Marcin Junczys-Dowmunt, Hany Hassan Awadalla, Alham Fikri Aji, Kenneth Heafield, Roman Grundkiewicz, Nikolay Bogoychev
  
  WNGT - EMNLP | November 2019
  
  Publication
- Time- and space-parallel simulation of air traffic networks
  
  Young Jin Kim, Dimitri Mavris, Richard Fujimoto
  
  SIMULATION (SAGE) | December 2019, 巻95(12): pp. 1213-1228
  
  Publication

- Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation
  
  Liliang Ren, Congcong Chen, Haoran Xu, Young Jin Kim, Adam Atkinson, Zheng Zhan, Jiankai Sun, Baolin Peng, Liyuan Liu, Shuohang Wang, Hao Cheng, Jianfeng Gao, Weizhu Chen, Yelong Shen
  
  NeurIPS 2025 | July 2025
  
  DOI
- SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation
  
  Zichong Li, Chen Liang, Zixuan Zhang, Ilgee Hong, Young Jin Kim, Weizhu Chen, Tuo Zhao
  
  COLM 2025 | July 2025
  
  Publication
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  ICML 2024 | January 2024
  
  DOI Publication Publication
- Task-Based MoE for Multitask Multilingual Machine Translation
  
  Hai Pham, Young Jin Kim, Subhabrata Mukherjee, David P Woodruff, Barnabas Poczos, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
  
  Haoran Xu, Young Jin Kim, Amr Sharaf, Hany Hassan Awadalla
  
  Accepted to ICLR 2024 | September 2023
  
  Preprint
- AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks Lakshmanan, V.S., Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL 2023 | June 2023
  
  Publication
- AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers
  
  Ganesh Jawahar, Subhabrata (Subho) Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Ahmed Awadallah, Sébastien Bubeck, Jianfeng Gao
  
  ACL | May 2023
  
  Publication プロジェクト
- Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2022
  
  Preprint
- Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU
  
  Hossam Amer, Young Jin Kim, Mohamed Afify, Hitokazu Matsushita, Hany Hassan Awadalla
  
  AMTA | August 2022
  
  Publication プロジェクト
- Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers
  
  Rui Liu, Young Jin Kim, Alexandre Muzio, Hany Hassan Awadalla
  
  ICML 2022 | July 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan Awadalla, Ruofei Zhang, Tuo Zhao, Jianfeng Gao
  
  ICLR 2022 | April 2022
  
  Publication プロジェクト
- Taming Sparsely Activated Transformer with Stochastic Experts.
  
  Simiao Zuo, Xiaodong Liu, Jian Jiao, Young Jin Kim, Hany Hassan, Ruofei Zhang, Tuo Zhao, Jianfeng Gao, Hany Hassan Awadalla
  
  October 2021
  
  Publication Publication
- FastFormers: Highly Efficient Transformer Models for Natural Language Understanding
  
  Young Jin Kim, Hany Hassan Awadalla
  
  SustaiNLP - EMNLP | November 2020
  
  Publication プロジェクト
- From Research to Production and Back: Ludicrously Fast Neural Machine Translation
  
  Young Jin Kim, Marcin Junczys-Dowmunt, Hany Hassan Awadalla, Alham Fikri Aji, Kenneth Heafield, Roman Grundkiewicz, Nikolay Bogoychev
  
  WNGT - EMNLP | November 2019
  
  Publication
- Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
  
  Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
  
  January 2024
  
  Preprint
- FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs
  
  Young Jin Kim, Rawn Henry, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Preprint
- Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness
  
  Young Jin Kim, Raffy Fahim, Hany Hassan Awadalla
  
  December 2023
  
  Publication
- PEMA: Plug-in External Memory Adaptation for Language Models
  
  HyunJin Kim, Young Jin Kim, JinYeong Bak
  
  November 2023
  
  Preprint
- How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation
  
  Amr Hendy, Mohamed Abdelrehim, Amr Sharaf, Vikas Raunak, Mohamed Gabr, Hitokazu Matsushita, Young Jin Kim, Mohamed Afify, Hany Hassan Awadalla
  
  February 2023
  
  Preprint
- Scalable and Efficient MoE Training for Multitask Multilingual Models
  
  Young Jin Kim, Ammar Ahmad Awan, Alexandre Muzio, Felipe Cruz Salinas, Liyang Lu, Amr Hendy, Samyam Rajbhandari, Yuxiong He, Hany Hassan Awadalla
  
  September 2021
  
  Publication Github プロジェクトプロジェクト
- Time- and space-parallel simulation of air traffic networks
  
  Young Jin Kim, Dimitri Mavris, Richard Fujimoto
  
  SIMULATION (SAGE) | December 2019, 巻95(12): pp. 1213-1228
  
  Publication

Young Jin Kim

出版物

Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation

SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs

Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness

Task-Based MoE for Multitask Multilingual Machine Translation

PEMA: Plug-in External Memory Adaptation for Language Models

A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models

AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation

AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers

How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation

Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production

Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU

Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers

Taming Sparsely Activated Transformer with Stochastic Experts

Taming Sparsely Activated Transformer with Stochastic Experts.

Scalable and Efficient MoE Training for Multitask Multilingual Models

FastFormers: Highly Efficient Transformer Models for Natural Language Understanding

Time- and space-parallel simulation of air traffic networks

From Research to Production and Back: Ludicrously Fast Neural Machine Translation

Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation

SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

Task-Based MoE for Multitask Multilingual Machine Translation

FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs

Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness

PEMA: Plug-in External Memory Adaptation for Language Models

A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models

AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation

AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers

How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation

Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production

Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers

Taming Sparsely Activated Transformer with Stochastic Experts

Taming Sparsely Activated Transformer with Stochastic Experts.

Scalable and Efficient MoE Training for Multitask Multilingual Models

FastFormers: Highly Efficient Transformer Models for Natural Language Understanding

From Research to Production and Back: Ludicrously Fast Neural Machine Translation

AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers

Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU

Taming Sparsely Activated Transformer with Stochastic Experts

Taming Sparsely Activated Transformer with Stochastic Experts.

Scalable and Efficient MoE Training for Multitask Multilingual Models

FastFormers: Highly Efficient Transformer Models for Natural Language Understanding

From Research to Production and Back: Ludicrously Fast Neural Machine Translation

Time- and space-parallel simulation of air traffic networks

Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation

SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

Task-Based MoE for Multitask Multilingual Machine Translation

A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models

AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation

AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers

Who Says Elephants Can’t Run: Bringing Large Scale MoE Models into Cloud Scale Production

Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU

Gating Dropout: Communication-efficient Regularization for Sparsely Activated Transformers

Taming Sparsely Activated Transformer with Stochastic Experts

Taming Sparsely Activated Transformer with Stochastic Experts.

FastFormers: Highly Efficient Transformer Models for Natural Language Understanding

From Research to Production and Back: Ludicrously Fast Neural Machine Translation

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs

Mixture of Quantized Experts (MoQE): Complementary Effect of Low-bit Quantization and Robustness

PEMA: Plug-in External Memory Adaptation for Language Models

How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation

Scalable and Efficient MoE Training for Multitask Multilingual Models

Time- and space-parallel simulation of air traffic networks