5 citations · 5 across the 10 of their papers we have counts for
11 papers
On a Separate Note: Robust Score-Informed Note Separation with a Two-Stream TFC-TDF U-Net and Adaptive Set Ownership
Benjamin Shiue-Hal Chou, Purvish Jajal, Nicholas John Eliopoulos +5
Score-informed note separation seeks to extract the performed waveform of all individual notes, often from a polyphonic recording. Existing deep learning systems generally only tar…
AdaPerceiver: Transformers with Adaptive Width, Depth, and Tokens
Purvish Jajal, Nick John Eliopoulos, Benjamin Shiue-Hal Chou +3
Modern transformer architectures achieve remarkable performance across tasks and domains but remain rigid in how they allocate computation at inference time. Real-world deployment…
LadderSym: A Multimodal Interleaved Transformer for Music Practice Error Detection
Benjamin Shiue-Hal Chou, Purvish Jajal, Nick John Eliopoulos +4
Music learners can greatly benefit from tools that accurately detect errors in their practice. Existing approaches typically compare audio recordings to music scores using heuristi…
Inference-Time Alignment of Diffusion Models via Evolutionary Algorithms
Purvish Jajal, Nick John Eliopoulos, Benjamin Shiue-Hal Chou +3
Diffusion models are state-of-the-art generative models, yet their samples often fail to satisfy application objectives such as safety constraints or domain-specific validity. Exis…
Detecting Music Performance Errors with Transformers
Benjamin Shiue-Hal Chou, Purvish Jajal, Nicholas John Eliopoulos +6
Beginner musicians often struggle to identify specific errors in their performances, such as playing incorrect notes or rhythms. There are two limitations in existing tools for mus…
Token Turing Machines are Efficient Vision Models
Purvish Jajal, Nick John Eliopoulos, Benjamin Shiue-Hal Chou +3
We propose Vision Token Turing Machines (ViTTM), an efficient, low-latency, memory-augmented Vision Transformer (ViT). Our approach builds on Neural Turing Machines and Token Turin…
Pruning One More Token is Enough: Leveraging Latency-Workload Non-Linearities for Vision Transformers on the Edge
Nick John Eliopoulos, Purvish Jajal, James C. Davis +3
This paper investigates how to efficiently deploy vision transformers on edge devices for small workloads. Recent methods reduce the latency of transformer neural networks by remov…
Reusing Deep Learning Models: Challenges and Directions in Software Engineering
James C. Davis, Purvish Jajal, Wenxin Jiang +3
Deep neural networks (DNNs) achieve state-of-the-art performance in many areas, including computer vision, system configuration, and question-answering. However, DNNs are expensive…
An automated approach for improving the inference latency and energy efficiency of pretrained CNNs by removing irrelevant pixels with focused convolutions
Caleb Tung, Nicholas Eliopoulos, Purvish Jajal +7
Computer vision often uses highly accurate Convolutional Neural Networks (CNNs), but these deep learning models are associated with ever-increasing energy and computation requireme…
PTMTorrent: A Dataset for Mining Open-source Pre-trained Model Packages
Wenxin Jiang, Nicholas Synovic, Purvish Jajal +5
Due to the cost of developing and training deep learning models from scratch, machine learning engineers have begun to reuse pre-trained models (PTMs) and fine-tune them for downst…
Analysis of Failures and Risks in Deep Learning Model Converters: A Case Study in the ONNX Ecosystem
Purvish Jajal, Wenxin Jiang, Arav Tewari +6
Software engineers develop, fine-tune, and deploy deep learning (DL) models using a variety of development frameworks and runtime environments. DL model converters move models betw…