no code implementations • 20 Apr 2018 • Emli-Mari Nel, Per Ola Kristensson, David J. C. MacKay
This paper presents the statistical model for Ticker [1], a novel probabilistic stereophonic single-switch text entry method for visually-impaired users with motor disabilities who rely on single-switch scanning systems to communicate.
no code implementations • 19 Sep 2017 • Shaona Ghosh, Per Ola Kristensson
Our models achieve a word level accuracy of $90\%$ and a character error rate CER of $2. 4\%$ over the Twitter typo dataset.
no code implementations • 27 May 2021 • Junxiao Shen, John Dudley, Per Ola Kristensson
Insufficient training data results in over-fitting and data augmentation is one approach to address this challenge.
no code implementations • 15 Apr 2022 • Liwei Chan, Yi-Chi Liao, George B. Mo, John J. Dudley, Chun-Lien Cheng, Per Ola Kristensson, Antti Oulasvirta
Designers reportedly struggle with design optimization tasks where they are asked to find a combination of design parameters that maximizes a given set of objectives.
no code implementations • 10 Aug 2023 • Junxiao Shen, John Dudley, Per Ola Kristensson
Additionally, in a user study, our system received a higher mean response score of 4. 13/5 compared to the human participants' score of 2. 46/5 on real-life episodic memory tasks.
no code implementations • 12 Oct 2023 • Junxiao Shen, John J. Dudley, Jingyao Zheng, Bill Byrne, Per Ola Kristensson
However, the task of prompting large language models to specialize in specific text prediction tasks can be challenging, particularly for designers without expertise in prompt engineering.
no code implementations • 20 Jan 2024 • Junxiao Shen, Matthias De Lange, Xuhai "Orson" Xu, Enmin Zhou, Ran Tan, Naveen Suda, Maciej Lazarewicz, Per Ola Kristensson, Amy Karlson, Evan Strasnick
We propose leveraging continual learning to make machine learning models adaptive to new tasks without degrading performance on previously learned tasks.