AIConfPaper
← Search

Mohammad Samragh

6 accepted papers

2026

Barriers for Learning in an Evolving World: Mathematical Understanding of Loss of Plasticity

Amir Joudaki, Giulia Lanzillotta, Mohammad Samragh, Seyed Iman Mirzadeh, Keivan Alizadeh, Thomas Hofmann +2
ICLR 2026poster

Deep learning models excel in stationary settings but suffer from loss of plasticity (LoP) in non-stationary environments. While prior literature characterizes LoP through symptoms like rank collapse of representations, it often lacks a mechanistic explanation for why gradient descent fails to recov…

Cited by 0Source ↗code ↗Scholar ↗
2025

SPD: Sync-Point Drop for Efficient Tensor Parallelism of Large Language Models

Han-Byul Kim, Duc N.M Hoang, Arnav Kundu, Mohammad Samragh, Minsik Cho
ICML 2025poster

With the rapid expansion in the scale of large language models (LLMs), enabling efficient distributed inference across multiple computing units has become increasingly critical. However, communication overheads from popular distributed inference techniques such as Tensor Parallelism pose a significa…

Cited by 0Source ↗PDF ↗Scholar ↗
2024

Improving Vision-Inspired Keyword Spotting Using Dynamic Module Skipping in Streaming Conformer Encoder

Alexandre Bittar, Paul Dixon, Mohammad Samragh, Kumari Nishu, Devang Naik
ICASSP 2024accepted

Using a vision-inspired keyword spotting framework, we propose an architecture with input-dependent dynamic depth capable of processing streaming audio. Specifically, we extend a conformer encoder with trainable binary gates that allow us to dynamically skip network modules according to the input au…

Cited by 0Source ↗Scholar ↗
2023

HEiMDaL: Highly Efficient Method for Detection and Localization of Wake-Words

Arnav Kundu, Mohammad Samragh, Minsik Cho, Priyanka Padmanabhan, Devang Naik
ICASSP 2023accepted

Streaming keyword spotting is a widely used solution for activating voice assistants. Methods based on Deep Neural Networks with Hidden Markov Model (DNN-HMM) have proven to be efficient and widely adopted in this space, primarily because of the ability to detect and identify the start and end of th…

Cited by 0Source ↗Scholar ↗
2023

I See What You Hear: A Vision-Inspired Method to Localize Words

Mohammad Samragh, Arnav Kundu, Ting-Yao Hu, Aman Chadha, Ashish Srivastava, Minsik Cho +2
ICASSP 2023accepted

This paper explores the possibility of using visual object detection techniques for word localization in speech data. Object detection has been thoroughly studied in the contemporary literature for visual data. Noting that an audio can be interpreted as a 1-dimensional image, object localization tec…

Cited by 0Source ↗Scholar ↗