Melvin Johnson

Computer Science · Physical Sciences

24h-index11.5kcitations57works179.02yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Natural Language Processing Techniques(52), Topic Modeling(47), Multimodal Machine Learning Applications(26), Speech Recognition and Synthesis(13), Text Readability and Simplification(8)

Publications57 total

Capabilities of Gemini Models in Medicine
arXiv (Cornell University)·2024· 88 citedOpen Access
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
arXiv (Cornell University)·2024· 270 citedOpen Access
Gemini: A Family of Highly Capable Multimodal Models
arXiv (Cornell University)·2023· 784 citedOpen Access
XTREME-UP: A User-Centric Scarce-Data Benchmark for Under-Represented Languages
arXiv (Cornell University)·2023· 3 citedOpen Access
PaLM 2 Technical Report
arXiv (Cornell University)·2023· 158 citedOpen Access
The unreasonable effectiveness of few-shot learning for machine translation
arXiv (Cornell University)·2023· 24 citedOpen Access
XTREME-UP: A User-Centric Scarce-Data Benchmark for Under-Represented Languages
2023· 10 citedOpen Access
Mu$^{2}$SLAM: Multitask, Multilingual Speech and Language Models
arXiv (Cornell University)·2022· 5 citedOpen Access
XTREME-S: Evaluating Cross-lingual Speech Representations
Interspeech 2022·2022· 14 cited
XTREME-S: Evaluating Cross-lingual Speech Representations
arXiv (Cornell University)·2022· 1 citedOpen Access
mSLAM: Massively multilingual joint pre-training for speech and text
arXiv (Cornell University)·2022· 59 citedOpen Access
DOCmT5: Document-Level Pretraining of Multilingual Language Models
Findings of the Association for Computational Linguistics: NAACL 2022·2022· 4 citedOpen Access
Multilingual Document-Level Translation Enables Zero-Shot Transfer From Sentences to Documents
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)·2022· 4 citedOpen Access
DOCmT5: Document-Level Pretraining of Multilingual Language Models
arXiv (Cornell University)·2021Open Access
SLAM: A Unified Encoder for Speech and Language Modeling via Speech-Text Joint Pre-Training
arXiv (Cornell University)·2021· 50 citedOpen Access
Multilingual Document-Level Translation Enables Zero-Shot Transfer From Sentences to Documents
arXiv (Cornell University)·2021· 2 citedOpen Access
HintedBT: Augmenting Back-Translation with Quality and Transliteration\n Hints
arXiv (Cornell University)·2021Open Access
HintedBT: Augmenting Back-Translation with Quality and Transliteration Hints
arXiv (Cornell University)·2021Open Access
MergeDistill: Merging Pre-trained Language Models using Distillation
arXiv (Cornell University)·2021· 1 citedOpen Access
nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?
arXiv (Cornell University)·2021· 1 citedOpen Access
page 1 of 3Next →

Frequent Co-authors

Orhan Fırat(23), Ankur Bapna(13), Wolfgang Macherey(12), Yonghui Wu(11), Maxim Krikun(10), Aditya Siddhant(9), Mike Schuster(8), Zhifeng Chen(8), Sebastian Ruder(8), Colin Cherry(8), Jason Riesa(7), Macduff Hughes(6), Jia Ye(6), Quoc V. Le(5), Yuan Cao(5), Greg S. Corrado(5), Yong Cheng(5), Junjie Hu(5), Graham Neubig(5), Ron J. Weiss(5)