Akarsh Kumar
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
ML×1AI×1
Frequent co-authors
Research Timeline
2026
Pretraining Recurrent Networks without Recurrence
This paper proposes Supervised Memory Training (SMT), a method for training nonlinear RNNs that sidesteps recurrent credit propagation entirely.
Highlighted terms show continued research focus across papers