← Search

Anirudh Sriram

2 accepted papers

2022

Input-specific Attention Subnetworks for Adversarial Detection

ACL 2022findings

Self-attention heads are characteristic of Transformer models and have been well studied for interpretability and pruning. In this work, we demonstrate an altogether different utility of attention heads, namely for adversarial detection. Specifically, we propose a method to construct input-specific…

2020

Joint Transformer/RNN Architecture for Gesture Typing in Indic Languages

COLING 2020main

Gesture typing is a method of typing words on a touch-based keyboard by creating a continuous trace passing through the relevant keys. This work is aimed at developing a keyboard that supports gesture typing in Indic languages. We begin by noting that when dealing with Indic languages, one needs to…