3 papers
cs.LG2026
Selecting Long-Horizon Trajectories for Reliable and Efficient Terminal-Agent Training
Cuong Dang, Hoang Anh Just, Ruoxi Jia
Terminal agents are commonly trained by imitating long teacher trajectories, yet how much of each trajectory to supervise remains unexplored. We study the \emph{supervision horizon…
cs.LG2026
Capacity-Dependent Effects of Data Selection for Reasoning
Cuong Dang, Hoang Anh Just, Ruoxi Jia
In reasoning supervised fine-tuning, candidate responses for the same instruction can differ substantially in how well they match the student's current distribution. Recent likelih…
math.NT2022
On Perfect Bases in Finite Abelian Groups
Bela Bajnok, Connor Berson, Hoang Anh Just
Let be a finite abelian group and be a positive integer. A subset of is called a {\em perfect -basis of } if each element of can be written uniquely as th…