177 citations · 308 across the 20 of their papers we have counts for
20 papers
Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +572
We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 t…
Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech
Siqi Ouyang, Shuoyang Ding, Oleksii Hrinchuk +4
Simultaneous speech translation (SST) generates translations while receiving partial speech input. Recent advances show that large language models (LLMs) can substantially improve…
Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aakshita Chandiramani +544
We describe the pre-training, post-training, and quantization of Nemotron 3 Super, a 120 billion (active 12 billion) parameter hybrid Mamba-Attention Mixture-of-Experts model. Nemo…
NVIDIA Nemotron 3: Efficient and Open Intelligence
NVIDIA, :, Aaron Blakeman +356
We introduce the Nemotron 3 family of models - Nano, Super, and Ultra. These models deliver strong agentic, reasoning, and conversational capabilities. The Nemotron 3 family uses a…
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +311
We present Nemotron 3 Nano 30B-A3B, a Mixture-of-Experts hybrid Mamba-Transformer language model. Nemotron 3 Nano was pretrained on 25 trillion text tokens, including more than 3 t…
NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
NVIDIA, :, Aarti Basant +214
We introduce Nemotron-Nano-9B-v2, a hybrid Mamba-Transformer language model designed to increase throughput for reasoning workloads while achieving state-of-the-art accuracy compar…
Anticipating Future with Large Language Model for Simultaneous Machine Translation
Siqi Ouyang, Oleksii Hrinchuk, Zhehuai Chen +4
Simultaneous machine translation (SMT) takes streaming input utterances and incrementally produces target text. Existing SMT methods mainly use the partial utterance that has alrea…
EMMeTT: Efficient Multimodal Machine Translation Training
Piotr Żelasko, Zhehuai Chen, Mengru Wang +7
A rising interest in the modality extension of foundation language models warrants discussion on the most effective, and efficient, multimodal training approach. This work focuses…
Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training for Enhanced Speech Recognition and Translation
Nithin Rao Koluguri, Travis Bartley, Hainan Xu +4
This paper presents a new method for training sequence-to-sequence models for speech recognition and translation tasks. Instead of the traditional approach of training models on sh…
Chain-of-Thought Prompting for Speech Translation
Ke Hu, Zhehuai Chen, Chao-Han Huck Yang +5
Large language models (LLMs) have demonstrated remarkable advancements in language understanding and generation. Building on the success of text-based LLMs, recent research has ada…
BESTOW: Efficient and Streamable Speech Language Model with the Best of Two Worlds in GPT and T5
Zhehuai Chen, He Huang, Oleksii Hrinchuk +5
Incorporating speech understanding capabilities into pretrained large-language models has become a vital research direction (SpeechLLM). The previous architectures can be categoriz…
SALM: Speech-augmented Language Model with In-context Learning for Speech Recognition and Translation
Zhehuai Chen, He Huang, Andrei Andrusenko +6
We present a novel Speech Augmented Language Model (SALM) with {\em multitask} and {\em in-context} learning capabilities. SALM comprises a frozen text LLM, a audio encoder, a moda…
Leveraging Synthetic Targets for Machine Translation
Sarthak Mittal, Oleksii Hrinchuk, Oleksii Kuchaiev
In this work, we provide a recipe for training machine translation models in a limited resource setting by leveraging synthetic target data generated using a large pre-trained mode…
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition
Dima Rekesh, Nithin Rao Koluguri, Samuel Kriman +8
Conformer-based models have become the dominant end-to-end architecture for speech processing tasks. With the objective of enhancing the conformer architecture for efficient traini…
Finding the Right Recipe for Low Resource Domain Adaptation in Neural Machine Translation
Virginia Adams, Sandeep Subramanian, Mike Chrzanowski +2
General translation models often still struggle to generate accurate translations in specialized domains. To guide machine translation practitioners and characterize the effectiven…
NVIDIA NeMo Neural Machine Translation Systems for English-German and English-Russian News and Biomedical Tasks at WMT21
Sandeep Subramanian, Oleksii Hrinchuk, Virginia Adams +1
This paper provides an overview of NVIDIA NeMo's neural machine translation systems for the constrained data track of the WMT21 News and Biomedical Shared Translation Tasks. Our ne…
Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
Somshubra Majumdar, Jagadeesh Balam, Oleksii Hrinchuk +3
We propose Citrinet - a new end-to-end convolutional Connectionist Temporal Classification (CTC) based automatic speech recognition (ASR) model. Citrinet is deep residual neural mo…
Correction of Automatic Speech Recognition with Transformer Sequence-to-sequence Model
Oleksii Hrinchuk, Mariya Popova, Boris Ginsburg
In this work, we introduce a simple yet efficient post-processing model for automatic speech recognition (ASR). Our model has Transformer-based encoder-decoder architecture which "…
NeMo: a toolkit for building AI applications using Neural Modules
Oleksii Kuchaiev, Jason Li, Huyen Nguyen +11
NeMo (Neural Modules) is a Python framework-agnostic toolkit for creating AI applications through re-usability, abstraction, and composition. NeMo is built around neural modules, c…
Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks
Boris Ginsburg, Patrice Castonguay, Oleksii Hrinchuk +7
We propose NovoGrad, an adaptive stochastic gradient descent method with layer-wise gradient normalization and decoupled weight decay. In our experiments on neural networks for ima…