Near-optimal P300 speller performance using large language models: A multi-model analysis with performance bounds
Nithin Parthasarathy, James Soetedjo, Saarang Panchavati, Nitya Parthasarathy, Dongwoo Lee, Corey Arnold, Nader Pouratian, William Speier · 2026
Read Near-optimal P300 speller performance using large language models: A multi-model analysis with performance bounds by Nithin Parthasarathy, James Soetedjo, Saarang Panchavati, Nitya Parthasarathy, Dongwoo Lee, Corey Arnold, Nader Pouratian, William Speier (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0349281.txt.
Description
Parthasarathy and seven co-authors published a 2026 PLOS ONE study on a brain-based typing tool. A P300 speller lets a person pick letters by focusing on flashes on a screen, while sensors catch the brain's response. The team tested large language models to predict the intended letters. They compared several models against formal performance bounds. Better spellers could give people who cannot move or speak a faster voice.
Catalog record
| CREATOR | Nithin Parthasarathy, James Soetedjo, Saarang Panchavati, Nitya Parthasarathy, Dongwoo Lee, Corey Arnold, Nader Pouratian, William Speier |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65374 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |