Výsledky vyhledávání - "Alexander M. Rush"

Interactive and Visual Prompt Engineering for Ad-hoc Task Adaptation With Large Language Models

Autor: Hendrik Strobelt, Albert Webson, Victor Sanh, Benjamin Hoover, Johanna Beyer, Hanspeter Pfister, Alexander M. Rush

Publikováno v: IEEE transactions on visualization and computer graphics.

State-of-the-art neural language models can now be used to solve ad-hoc language tasks through zero-shot prompting without the need for supervised training. This approach has gained popularity in recent years, and researchers have demonstrated prompt

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0071e18959c40d0d6cc310b323513c54
https://pubmed.ncbi.nlm.nih.gov/36191099

Zobrazit plný text záznamu

End-to-end learning of multiple sequence alignments with differentiable Smith-Waterman

Autor: Alexander M. Rush, Neil Thomas, Juannan Zhou, Justas Dauparas, Nicholas Bhattacharya, Peter K. Koo, Roshan Rao, Samantha Petti, Sergey Ovchinnikov

Publikováno v: Bioinformatics (Oxford, England). 39(1)

Motivation Multiple sequence alignments (MSAs) of homologous sequences contain information on structural and functional constraints and their evolutionary histories. Despite their importance for many downstream tasks, such as structure prediction, MS

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::51f972972b0b5a02243134a27ce8b202
https://pubmed.ncbi.nlm.nih.gov/36355460

Zobrazit plný text záznamu

Developmental Stage Classification of Embryos Using Two-Stream Neural Network with Linear-Chain Conditional Random Field

Autor: Daniel J. Needleman, Yoon Kim, Won-Dong Jang, Robbert Struyven, Helen Y. Yang, Brian Leahy, Alexander M. Rush, Donglai Wei, Hanspeter Pfister, Stanislav Lukyanenko, Dalit Ben-Yosef

Publikováno v: Med Image Comput Comput Assist Interv
Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 ISBN: 9783030872366
MICCAI (8)

The developmental process of embryos follows a monotonic order. An embryo can progressively cleave from one cell to multiple cells and finally transform to morula and blastocyst. For time-lapse videos of embryos, most existing developmental stage cla

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::983d1fd090eac5d4f2907b41e6f5a15e
https://pubmed.ncbi.nlm.nih.gov/34671767

Zobrazit plný text záznamu

GenNI: Human-AI Collaboration for Data-Backed Text Generation

Autor: Robert Krueger, Hanspeter Pfister, Alexander M. Rush, Jambay Kinley, Johanna Beyer, Hendrik Strobelt

Table2Text systems generate textual output based on structured data utilizing machine learning. These systems are essential for fluent natural language interfaces in tools such as virtual assistants; however, left to generate freely these ML systems

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::681e28135b9775deab19e79983ce9050
http://arxiv.org/abs/2110.10185

Zobrazit plný text záznamu

SM6: A 16nm System-on-Chip for Accurate and Noise-Robust Attention-Based NLP Applications : The 33rd Hot Chips Symposium – August 22-24, 2021

Autor: En-Yu Yang, David Brooks, Alexander M. Rush, Marco Donato, Coleman Hooper, Yuji Chai, Glenn G. Ko, Paul N. Whatmough, Gu-Yeon Wei, Thierry Tambe

Publikováno v: HCS

In this work, we present SM6, an SoC architecture for real-time denoised speech and NLP pipelines, featuring (1) MSSE: an unsupervised probabilistic sound source separation accelerator, (2) FlexNLP: a programmable inference accelerator for attention-

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::654237ef7679dcb15992c7daf7e7f34d
https://doi.org/10.1109/hcs52781.2021.9567180

Zobrazit plný text záznamu

Seq2seq-Vis: A Visual Debugging Tool for Sequence-to-Sequence Models

Autor: Hanspeter Pfister, Hendrik Strobelt, Michael Behrisch, Alexander M. Rush, Sebastian Gehrmann, Adam Perer

Publikováno v: IEEE Transactions on Visualization and Computer Graphics. 25:353-363

Neural Sequence-to-Sequence models have proven to be accurate and robust for many sequence prediction tasks, and have become the standard approach for automatic translation of text. The models work in a five stage blackbox process that involves encod

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9ddf11ba9d7d8016b7daf4b230c2151c
https://doi.org/10.1109/tvcg.2018.2865044

Zobrazit plný text záznamu

Low-Complexity Probing via Finding Subnetworks

Autor: Alexander M. Rush, Steven Cao, Victor Sanh

Publikováno v: NAACL-HLT

The dominant approach in probing neural networks for linguistic properties is to train a new shallow multi-layer perceptron (MLP) on top of the model's internal representations. This approach can detect properties encoded in the model, but at the cos

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d087da575b454ecbfbc500b3416dc6b3
http://arxiv.org/abs/2104.03514

Zobrazit plný text záznamu

9.8 A 25mm2 SoC for IoT Devices with 18ms Noise-Robust Speech-to-Text Latency via Bayesian Speech Denoising and Attention-Based Sequence-to-Sequence DNN Speech Recognition in 16nm FinFET

Autor: Gu-Yeon Wei, Coleman Hooper, Glenn G. Ko, Thierry Tambe, Paul N. Whatmough, Alexander M. Rush, En-Yu Yang, Marco Donato, David Brooks, Yuji Chai

Publikováno v: ISSCC

Automatic speech recognition (ASR) using deep learning is essential for user interfaces on IoT devices. However, previously published ASR chips [4 –7] do not consider realistic operating conditions, which are typically noisy and may include more th

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::fea2e13800304170161d2fbb93e1eac6
https://doi.org/10.1109/isscc42613.2021.9366062

Zobrazit plný text záznamu

Template Filling with Generative Transformers

Autor: Alexander M. Rush, Xinya Du, Claire Cardie

Publikováno v: NAACL-HLT

Template filling is generally tackled by a pipeline of two separate supervised systems – one for role-filler extraction and another for template/event recognition. Since pipelines consider events in isolation, they can suffer from error propagation

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::98430d10fb407af4d90f897003a5ad6f
https://doi.org/10.18653/v1/2021.naacl-main.70

Zobrazit plný text záznamu

Parameter-Efficient Transfer Learning with Diff Pruning

Autor: Demi Guo, Alexander M. Rush, Yoon Kim

Publikováno v: ACL/IJCNLP (1)

While task-specific finetuning of pretrained networks has led to significant empirical advances in NLP, the large size of networks makes finetuning difficult to deploy in multi-task, memory-constrained settings. We propose diff pruning as a simple ap

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e3239640b456ac0df0ddfdff22d702bb
https://doi.org/10.18653/v1/2021.acl-long.378

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání