papers

Publications (13)

cs.LG2026

KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning

Kris Shengjun Dong, Sahil Modi, Dima Nikiforov +4

Optimizing CUDA code across multiple generations of GPU architectures is challenging, as achieving peak performance requires an extensive exploration of an increasingly complex, ha…

eess.AS2023

Code-Switching Text Generation and Injection in Mandarin-English ASR

Haibin Yu, Yuxuan Hu, Yao Qian +7

Code-switching speech refers to a means of expression by mixing two or more languages within a single utterance. Automatic Speech Recognition (ASR) with End-to-End (E2E) modeling f…

cs.CL2023

Building High-accuracy Multilingual ASR with Gated Language Experts and Curriculum Training

Eric Sun, Jinyu Li, Yuxuan Hu +8

We propose gated language experts and curriculum training to enhance multilingual transformer transducer models without requiring language identification (LID) input from users dur…

cs.LG2026

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…

cs.CL2025

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…

cs.CL2026

Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

NVIDIA, :, Aaron Blakeman +571

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…

cs.CL2025

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…

astro-ph.IM2012

The SED Machine: A Spectrograph to Efficiently Classify Transient Events Discovered by PTF

Chow-Choong Ngeow, Nick Konidaris, Robert Quimby +4

The Palomar Transient Factory (PTF) is a project aimed to discover transients in the Universe, including Type Ia supernovae, core-collapse supernovae, and other exotic and rare tra…

cs.CL2023

SoftCorrect: Error Correction with Soft Detection for Automatic Speech Recognition

Yichong Leng, Xu Tan, Wenjie Liu +6

Error correction in automatic speech recognition (ASR) aims to correct those incorrect words in sentences generated by ASR models. Since recent ASR models usually have low word err…

cs.LG2026

SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits

Edward Lin, Sahil Modi, Siva Kumar Sastry Hari +30

As agentic AI systems become increasingly capable of generating and optimizing GPU kernels, progress is constrained by benchmarks that reward speedup over software baselines rather…

cs.CL2022

Mask the Correct Tokens: An Embarrassingly Simple Approach for Error Correction

Kai Shen, Yichong Leng, Xu Tan +4

Text error correction aims to correct the errors in text sequences such as those typed by humans or generated by speech recognition models. Previous error correction methods usuall…

cs.CL2022

FastCorrect 2: Fast Error Correction on Multiple Candidates for Automatic Speech Recognition

Yichong Leng, Xu Tan, Rui Wang +8

Error correction is widely used in automatic speech recognition (ASR) to post-process the generated sentence, and can further reduce the word error rate (WER). Although multiple ca…

eess.AS2023

FoundationTTS: Text-to-Speech for ASR Customization with Generative Language Model

Ruiqing Xue, Yanqing Liu, Lei He +4

Neural text-to-speech (TTS) generally consists of cascaded architecture with separately optimized acoustic model and vocoder, or end-to-end architecture with continuous mel-spectro…