Findings
Each post states one thing the lab learned, shows a figure from the paper, and cites it.
-
A language model scoring student peer feedback struggled on the same rubric criteria that human raters found hardest
Applying a four-part feedback-quality rubric to 295 peer comments, an open-source language model agreed well with researchers on whether a comment named a problem or suggested a fix, and poorly on the two criteria where the researchers also agreed least.
-
An open-model workflow for thematic analysis found nearly all the themes planted in synthetic test data
When the lab planted known themes in three synthetic datasets, the GATOS workflow produced a good match for 166 of 187 planted sub-themes, a partial match for 18 more, and missed 3.
-
Every reason people gave for leaving computer science showed up at every stage, from degree program to industry job
In 263 Reddit posts about leaving computer science, all six reasons the study identified appeared at all four points of departure, from students leaving a degree program to professionals leaving computing jobs, though the mix shifted by stage.
-
Engineering students drawn to nonprofits and to government want different things from a career
In a national survey of 4,605 final-year engineering students, interest in nonprofit work went with valuing volunteering and helping others, interest in government work went with wanting to tackle specific problems, and valuing job security pointed in opposite directions for the two.