3 papers
cs.LG2026
ALPHABET: A Laplace-Pole History Aggregator with Banked Exponential Transport
Daehwa Ko, JaeHyeon Kim, Oh Seong Kwon +1
Can a sequence model remain competitive with only a few thousand parameters and an explicitly auditable prediction interface? We introduce ALPHABET, a compact linear-time model tha…
cs.LG2026
TESLA: Taylor Expansion of Sinusoidal Learnable Activations
Daehwa Ko, Jaehyeon Kim, Seunghyun Ham +1
The parity problem--deciding whether the number of ones in a binary vector is odd or even--remains challenging for standard neural networks due to linear inseparability and the nee…
cs.CV2025
Amnesia as a Catalyst for Enhancing Black Box Pixel Attacks in Image Classification and Object Detection
Dongsu Song, Daehwa Ko, Jay Hoon Jung
It is well known that query-based attacks tend to have relatively higher success rates in adversarial black-box attacks. While research on black-box attacks is actively being condu…