Encoder-Free ECG-Language Models
William Han, Tony Chen, Chaojing Duan, Xiaoyu Song, Yihang Yao, Yuzhe Yang, Michael A. Rosenberg, Emerson Liu, Ding Zhao · 2026
Read Encoder-Free ECG-Language Models by William Han, Tony Chen, Chaojing Duan, Xiaoyu Song, Yihang Yao, Yuzhe Yang, Michael A. Rosenberg, Emerson Liu, Ding Zhao (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/arxiv-2601-18798.txt.
Description
William Han and eight colleagues drop a whole part from tools that read heart wave charts with the help of text. Such tools pair a chart of the heart's beats with a text model that can talk about it. Past tools first run the chart through a trained coder that turns it into a code. That coder adds bulk and cost to the build and to each train run.
Their tool, named ELF, swaps the coder for a lone thin sheet that maps the chart straight to the text model's space. The sheet trains with the model as one. A lean build could make heart chart tools cheap to train and swift to run. It could also bring them to small clinics with small kit. The team test if the lean form can match the grades of the full form.
Catalog record
| CREATOR | William Han, Tony Chen, Chaojing Duan, Xiaoyu Song, Yihang Yao, Yuzhe Yang, Michael A. Rosenberg, Emerson Liu, Ding Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-18798