Changye Li
@changyeli.bsky.social
Postdoc @ UW. NLP & speech processing. Thoughts are my own.
changyeli.github.io
changyeli.github.io
Hollow Knight spotted!
October 31, 2025 at 10:33 PM
Hollow Knight spotted!
I cried so hard reading this book 😢
September 30, 2025 at 6:32 PM
I cried so hard reading this book 😢
Don't forget snow tires!
September 15, 2025 at 8:26 PM
Don't forget snow tires!
Clearly people have different definitions on those terms 🤔
June 9, 2025 at 8:33 PM
Clearly people have different definitions on those terms 🤔
Please stop by if you are interested and happy to chat during the conference!
April 29, 2025 at 11:23 PM
Please stop by if you are interested and happy to chat during the conference!
and b) a poster presentation on May 3rd (yes, the same day) between 2:50pm to 3:30pm on the better performance of smaller language models for detecting formal thought disorder (aclanthology.org/2025.clpsych...) at Tesuque/Zuni Suite.
Bigger But Not Better: Small Neural Language Models Outperform LLMs in Detection of Thought Disorder
Changye Li, Weizhe Xu, Serguei Pakhomov, Ellen Bradley, Dror Ben-Zeev, Trevor Cohen. Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology (CLPsych 2025). 2025.
aclanthology.org
April 29, 2025 at 11:23 PM
and b) a poster presentation on May 3rd (yes, the same day) between 2:50pm to 3:30pm on the better performance of smaller language models for detecting formal thought disorder (aclanthology.org/2025.clpsych...) at Tesuque/Zuni Suite.
I use the combination of semantic scholar, scholar QA, and perplexity. But they all need to double check with the returned results - their summarization could be "not right" on some key points.
April 24, 2025 at 10:21 PM
I use the combination of semantic scholar, scholar QA, and perplexity. But they all need to double check with the returned results - their summarization could be "not right" on some key points.
Find out more in our preprint: arxiv.org/pdf/2503.20104 [2/2]
arxiv.org
March 31, 2025 at 6:48 PM
Find out more in our preprint: arxiv.org/pdf/2503.20104 [2/2]
You can find preprint here: arxiv.org/abs/2503.20103
Bigger But Not Better: Small Neural Language Models Outperform Large Language Models in Detection of Thought Disorder
Disorganized thinking is a key diagnostic indicator of schizophrenia-spectrum disorders. Recently, clinical estimates of the severity of disorganized thinking have been shown to correlate with measure...
arxiv.org
March 28, 2025 at 7:47 PM
You can find preprint here: arxiv.org/abs/2503.20103