Publications | Mohamed S. Abdelfattah

2025

preprint

SplitReason: Learning To Offload Reasoning

Yash Akhauri, Anthony Fei, Chi-Chih Chang, Ahmed F. AbouElhamayed, Yueying Li, Mohamed S. Abdelfattah

@inproceedings{0,
  title = {SplitReason: Learning To Offload Reasoning},
  author = {Akhauri, Yash and Fei, Anthony and Chang, Chi-Chih and AbouElhamayed, Ahmed F. and Li, Yueying and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/abs/2504.16379}
}

preprint

FlashDepth: Real-time Streaming Depth Estimation at 2K Resolution

Gene Chou, Wenqi Xian, Guandao Yang, Mohamed S. Abdelfattah, Bharath Hariharan, Noah Snavely, Ning Yu, Paul Debevec

@inproceedings{1,
  title = {FlashDepth: Real-time Streaming Depth Estimation at 2K Resolution},
  author = {Chou, Gene and Xian, Wenqi and Yang, Guandao and Abdelfattah, Mohamed S. and Hariharan, Bharath and Snavely, Noah and Yu, Ning and Debevec, Paul},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2504.07093}
}

preprint

Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion

Zhanqiu Hu, Jian Meng, Yash Akhauri, Mohamed S. Abdelfattah, Jae-sun Seo, Zhiru Zhang, Udit Gupta

@inproceedings{2,
  title = {Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion},
  author = {Hu, Zhanqiu and Meng, Jian and Akhauri, Yash and Abdelfattah, Mohamed S. and Seo, Jae-sun and Zhang, Zhiru and Gupta, Udit},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2505.21467}
}

preprint

The Power of Negative Zero: Datatype Customization for Quantized Large Language Models

Yuzong Chen, Xilai Dai, Chi-chih Chang, Yash Akhauri, Mohamed S. Abdelfattah

@inproceedings{3,
  title = {The Power of Negative Zero: Datatype Customization for Quantized Large Language Models},
  author = {Chen, Yuzong and Dai, Xilai and Chang, Chi-chih and Akhauri, Yash and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2501.04052}
}

preprint

xKV: Cross-Layer SVD for KV-Cache Compression

Chi-Chih Chang, Chien-Yu Lin, Yash Akhauri, Wei-Cheng Lin, Kai-Chiang Wu, Luis Ceze, Mohamed S. Abdelfattah

@inproceedings{4,
  title = {xKV: Cross-Layer SVD for KV-Cache Compression},
  author = {Chang, Chi-Chih and Lin, Chien-Yu and Akhauri, Yash and Lin, Wei-Cheng and Wu, Kai-Chiang and Ceze, Luis and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2503.18893}
}

preprint

TokenButler: Token Importance is Predictable

Yash Akhauri, Safeen Huda, Mohamed S. Abdelfattah

@inproceedings{6,
  title = {TokenButler: Token Importance is Predictable},
  author = {Akhauri, Yash and Huda, Safeen and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2503.07518}
}

preprint

SparAMX: Accelerating Compressed LLMs Token Generation on AMX-powered CPUs

Ahmed F. AbouElhamayed, Jordan Dotzel, Yash Akhauri, Chi-Chih Chang, Sameh Gobriel, J. Pablo Muñoz, Vui Seng Chua, Nilesh Jain, Mohamed S. Abdelfattah

@inproceedings{7,
  title = {SparAMX: Accelerating Compressed LLMs Token Generation on AMX-powered CPUs},
  author = {AbouElhamayed, Ahmed F. and Dotzel, Jordan and Akhauri, Yash and Chang, Chi-Chih and Gobriel, Sameh and Muñoz, J. Pablo and Chua, Vui Seng and Jain, Nilesh and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2502.12444}
}

ICML

Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space Models

Hung-Yueh Chiang, Chi-Chih Chang, Natalia Frumkin, Kai-Chiang Wu, Mohamed S. Abdelfattah, Diana Marculescu

International Conference on Machine Learning

@inproceedings{8,
  title = {Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space Models},
  author = {Chiang, Hung-Yueh and Chang, Chi-Chih and Frumkin, Natalia and Wu, Kai-Chiang and Abdelfattah, Mohamed S. and Marculescu, Diana},
  booktitle = {International Conference on Machine Learning},
  abbr = {ICML},
  year = {2025},
  pdf = { https://arxiv.org/pdf/2503.22879}
}

ICLR

Palu: Compressing KV-Cache with Low-Rank Projection

Chi-Chih Chang, Wei-Cheng Lin, Chien-Yu Lin, Chong-Yan Chen, Yu-Fang Hu, Pei-Shuo Wang, Ning-Chi Huang, Luis Ceze, Mohamed S. Abdelfattah, Kai-Chiang Wu

International Conference on Learning Representations

@inproceedings{9,
  title = {Palu: Compressing KV-Cache with Low-Rank Projection},
  author = {Chang, Chi-Chih and Lin, Wei-Cheng and Lin, Chien-Yu and Chen, Chong-Yan and Hu, Yu-Fang and Wang, Pei-Shuo and Huang, Ning-Chi and Ceze, Luis and Abdelfattah, Mohamed S. and Wu, Kai-Chiang},
  booktitle = {International Conference on Learning Representations},
  abbr = {ICLR},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2407.21118}
}

HPCA

BitMoD: Bit-serial Mixture-of-Datatype LLM Acceleration

Yuzong Chen, Ahmed F. AbouElhamayed, Xilai Dai, Yang Wang, Marta Andronic, George A. Constantinides, Mohamed S. Abdelfattah

IEEE International Symposium on High-Performance Computer Architecture

@inproceedings{10,
  title = {BitMoD: Bit-serial Mixture-of-Datatype LLM Acceleration},
  author = {Chen, Yuzong and AbouElhamayed, Ahmed F. and Dai, Xilai and Wang, Yang and Andronic, Marta and Constantinides, George A. and Abdelfattah, Mohamed S.},
  booktitle = {IEEE International Symposium on High-Performance Computer Architecture},
  abbr = {HPCA},
  year = {2025},
  pdf = {https://arxiv.org/pdf/2411.11745}
}

2024

preprint

Attamba: Attending To Multi-Token States

Yash Akhauri, Ahmed F. AbouElhamayed, Yifei Gao, Chi-Chih Chang, Nilesh Jain, Mohamed S. Abdelfattah

@inproceedings{5,
  title = {Attamba: Attending To Multi-Token States},
  author = {Akhauri, Yash and AbouElhamayed, Ahmed F. and Gao, Yifei and Chang, Chi-Chih and Jain, Nilesh and Abdelfattah, Mohamed S.},
  booktitle = {},
  abbr = {preprint},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2411.17685}
}

EMNLP

ShadowLLM: Predictor-based Contextual Sparsity for Large Language Models

Yash Akhauri, Ahmed F. AbouElhamayed, Jordan Dotzel, Zhiru Zhang, Alexander M. Rush, Safeen Huda, Mohamed S. Abdelfattah

Conference on Empirical Methods in Natural Language Processing

@inproceedings{11,
  title = {ShadowLLM: Predictor-based Contextual Sparsity for Large Language Models},
  author = {Akhauri, Yash and AbouElhamayed, Ahmed F. and Dotzel, Jordan and Zhang, Zhiru and Rush, Alexander M. and Huda, Safeen and Abdelfattah, Mohamed S.},
  booktitle = {Conference on Empirical Methods in Natural Language Processing},
  abbr = {EMNLP},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2406.16635}
}

MICRO

BBS: Bi-directional Bit-level Sparsity for Deep Learning Acceleration

Yuzong Chen, Jian Meng, Jae-sun Seo, Mohamed S. Abdelfattah

IEEE/ACM International Symposium on Microarchitecture

@inproceedings{12,
  title = {BBS: Bi-directional Bit-level Sparsity for Deep Learning Acceleration},
  author = {Chen, Yuzong and Meng, Jian and Seo, Jae-sun and Abdelfattah, Mohamed S.},
  booktitle = {IEEE/ACM International Symposium on Microarchitecture},
  abbr = {MICRO},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2409.05227}
}

AutoML

FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search

Jordan Dotzel, Gang Wu, Andrew Li, Muhammad Umar, Yun Ni, Mohamed S. Abdelfattah, Zhiru Zhang, Liqun Cheng, Martin G. Dixon, Norman P. Jouppi, Quoc V. Le, Sheng Li

International Conference on Automated Machine Learning

Best Paper Award

@inproceedings{13,
  title = {FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search},
  author = {Dotzel, Jordan and Wu, Gang and Li, Andrew and Umar, Muhammad and Ni, Yun and Abdelfattah, Mohamed S. and Zhang, Zhiru and Cheng, Liqun and Dixon, Martin G. and Jouppi, Norman P. and Le, Quoc V. and Li, Sheng},
  booktitle = {International Conference on Automated Machine Learning},
  abbr = {AutoML},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2308.03290},
  award = {Best Paper Award}
}

FPL

Kratos: An FPGA Benchmark for Unrolled DNNs with Fine-Grained Sparsity and Mixed Precision

Xilai Dai, Yuzong Chen, Mohamed S. Abdelfattah

International Conference on Field-Programmable Logic and Applications

@inproceedings{14,
  title = {Kratos: An FPGA Benchmark for Unrolled DNNs with Fine-Grained Sparsity and Mixed Precision},
  author = {Dai, Xilai and Chen, Yuzong and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Field-Programmable Logic and Applications},
  abbr = {FPL},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2407.06033}
}

ICML

Encodings for Prediction-based Neural Architecture Search

Yash Akhauri, Mohamed S. Abdelfattah

International Conference on Machine Learning

@inproceedings{15,
  title = {Encodings for Prediction-based Neural Architecture Search},
  author = {Akhauri, Yash and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Machine Learning},
  abbr = {ICML},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2403.02484}
}

ICML

Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs

Jordan Dotzel, Yuzong Chen, Bahaa Kotb, Sushma Prasad, Gang Wu, Sheng Li, Mohamed S. Abdelfattah, Zhiru Zhang

International Conference on Machine Learning

@inproceedings{16,
  title = {Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs},
  author = {Dotzel, Jordan and Chen, Yuzong and Kotb, Bahaa and Prasad, Sushma and Wu, Gang and Li, Sheng and Abdelfattah, Mohamed S. and Zhang, Zhiru},
  booktitle = {International Conference on Machine Learning},
  abbr = {ICML},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2405.03103}
}

ICML

Towards Neural Architecture Search through Hierarchical Generative Modeling

Lichuan Xiang, Łukasz Dudziak, Mohamed S. Abdelfattah, Abhinav Mehrotra, Nicholas D. Lane, Hongkai Wen

International Conference on Machine Learning

@inproceedings{17,
  title = {Towards Neural Architecture Search through Hierarchical Generative Modeling},
  author = {Xiang, Lichuan and Dudziak, Łukasz and Abdelfattah, Mohamed S. and Mehrotra, Abhinav and Lane, Nicholas D. and Wen, Hongkai},
  booktitle = {International Conference on Machine Learning},
  abbr = {ICML},
  year = {2024},
  pdf = {https://openreview.net/pdf?id=VdZfEMuoj2}
}

DAC

Beyond Inference: Performance Analysis of DNN Server Overheads for Computer Vision

Ahmed F. AbouElhamayed, Susanne Balle, Deshanand Singh, Mohamed S. Abdelfattah

Design Automation Conference

@inproceedings{18,
  title = {Beyond Inference: Performance Analysis of DNN Server Overheads for Computer Vision},
  author = {AbouElhamayed, Ahmed F. and Balle, Susanne and Singh, Deshanand and Abdelfattah, Mohamed S.},
  booktitle = {Design Automation Conference},
  abbr = {DAC},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2403.12981.pdf}
}

MLSYS

On Latency Predictors for Neural Architecture Search

Yash Akhauri, Mohamed S. Abdelfattah

International Conference on Machine Learning and Systems

@inproceedings{19,
  title = {On Latency Predictors for Neural Architecture Search},
  author = {Akhauri, Yash and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Machine Learning and Systems},
  abbr = {MLSYS},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2403.02446.pdf}
}

TRETS

PQA: Exploring the Potential of Product Quantization in DNN Hardware Acceleration

Ahmed F. AbouElhamayed, Angela Cui, Javier Fernandez-Marques, Nicholas D. Lane, Mohamed S. Abdelfattah

ACM Transactions on Reconfigurable Technology and Systems

FCCM’24 Journal Track.

@article{20,
  title = {PQA: Exploring the Potential of Product Quantization in DNN Hardware Acceleration},
  author = {AbouElhamayed, Ahmed F. and Cui, Angela and Fernandez-Marques, Javier and Lane, Nicholas D. and Abdelfattah, Mohamed S.},
  journal = {ACM Transactions on Reconfigurable Technology and Systems},
  abbr = {TRETS},
  year = {2024},
  pdf = {https://arxiv.org/pdf/2305.18334},
  award = {FCCM'24 Journal Track.}
}

2023

FPT

M4BRAM: Mixed-Precision Matrix-Matrix Multiplication in FPGA Block RAMs

Yuzong Chen, Jordan Dotzel, Mohamed S. Abdelfattah

International Conference on Field-Programmable Technology

@inproceedings{21,
  title = {M4BRAM: Mixed-Precision Matrix-Matrix Multiplication in FPGA Block RAMs},
  author = {Chen, Yuzong and Dotzel, Jordan and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Field-Programmable Technology},
  abbr = {FPT},
  year = {2023},
  pdf = {https://arxiv.org/pdf/2311.02758.pdf}
}

ICCAD

DiviML: A Module-based Heuristic for Mapping Neural Networks onto Heterogeneous Platforms

Yassine Ghannane, Mohamed S. Abdelfattah

International Conference on Computer-Aided Design

@inproceedings{22,
  title = {DiviML: A Module-based Heuristic for Mapping Neural Networks onto Heterogeneous Platforms},
  author = {Ghannane, Yassine and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Computer-Aided Design},
  abbr = {ICCAD},
  year = {2023},
  pdf = {https://arxiv.org/pdf/2308.00127.pdf}
}

AutoML

Multi-Predict: Few Shot Predictors for Efficient Neural Architecture Search

Yash Akhauri, Mohamed S. Abdelfattah

International Conference on Automated Machine Learning

@inproceedings{23,
  title = {Multi-Predict: Few Shot Predictors for Efficient Neural Architecture Search},
  author = {Akhauri, Yash and Abdelfattah, Mohamed S.},
  booktitle = {International Conference on Automated Machine Learning},
  abbr = {AutoML},
  year = {2023},
  pdf = {https://arxiv.org/pdf/2306.02459.pdf}
}

FCCM

BRAMAC: Compute-in-BRAM Architectures for Multiply-Accumulate on FPGAs

Yuzong Chen, Mohamed S. Abdelfattah

International Symposium On Field-Programmable Custom Computing Machines

@inproceedings{24,
  title = {BRAMAC: Compute-in-BRAM Architectures for Multiply-Accumulate on FPGAs},
  author = {Chen, Yuzong and Abdelfattah, Mohamed S.},
  booktitle = {International Symposium On Field-Programmable Custom Computing Machines},
  abbr = {FCCM},
  year = {2023},
  pdf = {https://arxiv.org/pdf/2304.03974.pdf}
}

TRETS

Learned Connectivity Sparsification for LUT-based Neural Networks

Erwei Wang, Georgios Stavrou, Peter Cheung, George Constantinides, Mohamed S. Abdelfattah, James Davis

ACM Transactions on Reconfigurable Technology and Systems

@article{25,
  title = {Learned Connectivity Sparsification for LUT-based Neural Networks},
  author = {Wang, Erwei and Stavrou, Georgios and Cheung, Peter and Constantinides, George and Abdelfattah, Mohamed S. and Davis, James},
  journal = {ACM Transactions on Reconfigurable Technology and Systems},
  abbr = {TRETS},
  year = {2023},
  pdf = {/assets/pdf/trets23.pdf}
}

AAAI

Zero-Cost Operation Scoring in Differentiable Architecture Search

Lichuan Xiang, Łukasz Dudziak, Mohamed S. Abdelfattah, Thomas Chau, Nicholas D. Lane, Hongkai Wen

AAAI Conference on Artificial Intelligence

@inproceedings{26,
  title = {Zero-Cost Operation Scoring in Differentiable Architecture Search},
  author = {Xiang, Lichuan and Dudziak, Łukasz and Abdelfattah, Mohamed S. and Chau, Thomas and Lane, Nicholas D. and Wen, Hongkai},
  booktitle = {AAAI Conference on Artificial Intelligence},
  abbr = {AAAI},
  year = {2023},
  pdf = {https://arxiv.org/pdf/2106.06799.pdf}
}

2022

NeurIPS

BLOX: Macro Neural Architecture Search Benchmark and Algorithms

Thomas Chau, Łukasz Dudziak, Hongkai Wen, Nicholas D. Lane, Mohamed S. Abdelfattah

Conference on Neural Information Processing Systems

@inproceedings{27,
  title = {BLOX: Macro Neural Architecture Search Benchmark and Algorithms},
  author = {Chau, Thomas and Dudziak, Łukasz and Wen, Hongkai and Lane, Nicholas D. and Abdelfattah, Mohamed S.},
  booktitle = {Conference on Neural Information Processing Systems},
  abbr = {NeurIPS},
  year = {2022},
  pdf = {https://arxiv.org/pdf/2210.07271.pdf}
}

MICRO

Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design

Hongxiang Fan, Thomas Chau, Stylianos Venieris, Royson Lee, Alexandros Kouris, Wayne Luk, Nicholas D. Lane, Mohamed S. Abdelfattah

IEEE/ACM International Symposium on Microarchitecture

@inproceedings{28,
  title = {Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design},
  author = {Fan, Hongxiang and Chau, Thomas and Venieris, Stylianos and Lee, Royson and Kouris, Alexandros and Luk, Wayne and Lane, Nicholas D. and Abdelfattah, Mohamed S.},
  booktitle = {IEEE/ACM International Symposium on Microarchitecture},
  abbr = {MICRO},
  year = {2022},
  pdf = {https://arxiv.org/pdf/2209.09570.pdf}
}

preprint

Zero-Cost Operation Scoring in Differentiable Architecture Search

Lichuan Xiang, Łukasz Dudziak, Mohamed S. Abdelfattah, Thomas Chau, Nicholas D. Lane, Hongkai Wen

@inproceedings{29,
  title = {Zero-Cost Operation Scoring in Differentiable Architecture Search},
  author = {Xiang, Lichuan and Dudziak, Łukasz and Abdelfattah, Mohamed S. and Chau, Thomas and Lane, Nicholas D. and Wen, Hongkai},
  booktitle = {},
  abbr = {preprint},
  year = {2022},
  pdf = {https://arxiv.org/pdf/2106.06799.pdf}
}

FPGA

Logic Shrinkage: Learned FPGA Netlist Sparsity for Efficient Neural Network Inference

Erwei Wang, James Davis, Georgios Stavrou, Peter Cheung, George Constantinides, Mohamed S. Abdelfattah

International Symposium on Field-Programmable Gate Arrays

Nominated for the best paper award.

@inproceedings{30,
  title = {Logic Shrinkage: Learned FPGA Netlist Sparsity for Efficient Neural Network Inference},
  author = {Wang, Erwei and Davis, James and Stavrou, Georgios and Cheung, Peter and Constantinides, George and Abdelfattah, Mohamed S.},
  booktitle = {International Symposium on Field-Programmable Gate Arrays},
  abbr = {FPGA},
  year = {2022},
  pdf = {https://arxiv.org/pdf/2112.02346.pdf},
  award = {Nominated for the best paper award.}
}

2021

ICLR

Zero-Cost Proxies for Lightweight NAS

Mohamed S. Abdelfattah, Abhinav Mehrotra, Łukasz Dudziak, Nicholas D. Lane

International Conference on Learning Representations

@inproceedings{31,
  title = {Zero-Cost Proxies for Lightweight NAS},
  author = {Abdelfattah, Mohamed S. and Mehrotra, Abhinav and Dudziak, Łukasz and Lane, Nicholas D.},
  booktitle = {International Conference on Learning Representations},
  abbr = {ICLR},
  year = {2021},
  pdf = {https://openreview.net/pdf?id=0cmMMy8J5q}
}

ICLR

NAS-Bench-ASR: Reproducible Neural Architecture Search for Speech Recognition

Abhinav Mehrotra, Alberto Gil C. P. Ramos, Sourav Bhattacharya, Łukasz Dudziak, Ravichander Vipperla, Thomas Chau, Mohamed S. Abdelfattah, Samin Ishtiaq, Nicholas D. Lane

International Conference on Learning Representations

@inproceedings{32,
  title = {NAS-Bench-ASR: Reproducible Neural Architecture Search for Speech Recognition},
  author = {Mehrotra, Abhinav and Ramos, Alberto Gil C. P. and Bhattacharya, Sourav and Dudziak, Łukasz and Vipperla, Ravichander and Chau, Thomas and Abdelfattah, Mohamed S. and Ishtiaq, Samin and Lane, Nicholas D.},
  booktitle = {International Conference on Learning Representations},
  abbr = {ICLR},
  year = {2021},
  pdf = {https://openreview.net/pdf?id=CU0APx9LMaL}
}

ICCV

Temporal Kernel Consistency for Blind Video Super-Resolution

Lichuan Chang, Royson Lee, Hongkai Wen, Mohamed S. Abdelfattah, Nicholas D. Lane

International Conference on Computer Vision Workshops

@inproceedings{33,
  title = {Temporal Kernel Consistency for Blind Video Super-Resolution},
  author = {Chang, Lichuan and Lee, Royson and Wen, Hongkai and Abdelfattah, Mohamed S. and Lane, Nicholas D.},
  booktitle = {International Conference on Computer Vision Workshops},
  abbr = {ICCV},
  year = {2021},
  pdf = {https://arxiv.org/pdf/2108.08305.pdf}
}

2020

NeurIPS

BRP-NAS: Prediction-based NAS using GCNs

Łukasz Dudziak, Thomas Chau, Mohamed S. Abdelfattah, Royson Lee, Hyeji Kim, Nicholas D. Lane

Conference on Neural Information Processing Systems

@inproceedings{34,
  title = {BRP-NAS: Prediction-based NAS using GCNs},
  author = {Dudziak, Łukasz and Chau, Thomas and Abdelfattah, Mohamed S. and Lee, Royson and Kim, Hyeji and Lane, Nicholas D.},
  booktitle = {Conference on Neural Information Processing Systems},
  abbr = {NeurIPS},
  year = {2020},
  pdf = {https://arxiv.org/pdf/2007.08668.pdf}
}

INTERSPEECH

Iterative Compression of End-to-End ASR Model Using Reinforcement learning

Abhinav Mehrotra, Łukasz Dudziak, Jinsu Yeo, Younyoon Lee, Ravichander Vipperla, Mohamed S. Abdelfattah, Sangjeong Lee, Daehyun Kim, Nicholas D. Lane

Conference of the International Speech Communication Association

@inproceedings{35,
  title = {Iterative Compression of End-to-End ASR Model Using Reinforcement learning},
  author = {Mehrotra, Abhinav and Dudziak, Łukasz and Yeo, Jinsu and Lee, Younyoon and Vipperla, Ravichander and Abdelfattah, Mohamed S. and Lee, Sangjeong and Kim, Daehyun and Lane, Nicholas D.},
  booktitle = {Conference of the International Speech Communication Association},
  abbr = {INTERSPEECH},
  year = {2020},
  pdf = {https://arxiv.org/pdf/2008.02897.pdf}
}

ECCV

Journey Towards Tiny Perceptual Super-Resolution

Royson Lee, Łukasz Dudziak, Mohamed S. Abdelfattah, Hyeji Kim, Stylianos Veneris, Hongkai Wen, Nicholas D. Lane

European Conference on Computer Vision

@inproceedings{36,
  title = {Journey Towards Tiny Perceptual Super-Resolution},
  author = {Lee, Royson and Dudziak, Łukasz and Abdelfattah, Mohamed S. and Kim, Hyeji and Veneris, Stylianos and Wen, Hongkai and Lane, Nicholas D.},
  booktitle = {European Conference on Computer Vision},
  abbr = {ECCV},
  year = {2020},
  pdf = {https://arxiv.org/pdf/2007.04356.pdf}
}

DAC

Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator

Mohamed S. Abdelfattah, Łukasz Dudziak, Thomas Chau, Royson Lee, Hyeji Kim, Nicholas D. Lane

Design Automation Conference

@inproceedings{37,
  title = {Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator},
  author = {Abdelfattah, Mohamed S. and Dudziak, Łukasz and Chau, Thomas and Lee, Royson and Kim, Hyeji and Lane, Nicholas D.},
  booktitle = {Design Automation Conference},
  abbr = {DAC},
  year = {2020},
  pdf = {https://arxiv.org/pdf/2002.05022.pdf}
}

2019

INTERSPEECH

ShrinkML: End-to-End ASR Model Compression Using Reinforcement Learning

Łukasz Dudziak, Mohamed S. Abdelfattah, Ravichander Vipperla, Stefanos Laskaridis, Nicholas D. Lane

Conference of the International Speech Communication Association

@inproceedings{39,
  title = {ShrinkML: End-to-End ASR Model Compression Using Reinforcement Learning},
  author = {Dudziak, Łukasz and Abdelfattah, Mohamed S. and Vipperla, Ravichander and Laskaridis, Stefanos and Lane, Nicholas D.},
  booktitle = {Conference of the International Speech Communication Association},
  abbr = {INTERSPEECH},
  year = {2019},
  pdf = {https://arxiv.org/pdf/1907.03540.pdf}
}

2018

FPL

DLA: Compiler and FPGA Overlay for Neural Network Inference Acceleration

Mohamed S. Abdelfattah, David Han, Andrew Bitar, Roberto DiCecco, Shane O’Connell, Nitika Shanker, Joseph Chu, Ian Prins, Joshua Fender, Andrew C. Ling, Gordon R. Chiu

International Conference on Field-Programmable Logic and Applications

@inproceedings{40,
  title = {DLA: Compiler and FPGA Overlay for Neural Network Inference Acceleration},
  author = {Abdelfattah, Mohamed S. and Han, David and Bitar, Andrew and DiCecco, Roberto and O’Connell, Shane and Shanker, Nitika and Chu, Joseph and Prins, Ian and Fender, Joshua and Ling, Andrew C. and Chiu, Gordon R.},
  booktitle = {International Conference on Field-Programmable Logic and Applications},
  abbr = {FPL},
  year = {2018},
  pdf = {https://arxiv.org/pdf/1807.06434.pdf}
}

HEART

Harnessing Numerical Flexibility for Deep Learning on FPGAs

Andrew C. Ling, Mohamed S. Abdelfattah, Shane O’Connell, Andrew Bitar, David Han, Roberto Dicecco, Suchit Subhaschandra, Chris N Johnson, Dmitry Denisenko, Josh Fender, Gordon R. Chiu

International Symposium on Highly-Efficient Accelerators and Reconfigurable Technologies

@inproceedings{41,
  title = {Harnessing Numerical Flexibility for Deep Learning on FPGAs},
  author = {Ling, Andrew C. and Abdelfattah, Mohamed S. and O'Connell, Shane and Bitar, Andrew and Han, David and Dicecco, Roberto and Subhaschandra, Suchit and Johnson, Chris N and Denisenko, Dmitry and Fender, Josh and Chiu, Gordon R.},
  booktitle = {International Symposium on Highly-Efficient Accelerators and Reconfigurable Technologies},
  abbr = {HEART},
  year = {2018},
  pdf = {/assets/pdf/heart18.pdf}
}

ISPD

Flexibility: FPGAs and CAD in Deep Learning Acceleration

Gordon R. Chiu, Andrew C. Ling, Davor Capalija, Andrew Bitar, Mohamed S. Abdelfattah

International Symposium on Physical Design

@inproceedings{42,
  title = {Flexibility: FPGAs and CAD in Deep Learning Acceleration},
  author = {Chiu, Gordon R. and Ling, Andrew C. and Capalija, Davor and Bitar, Andrew and Abdelfattah, Mohamed S.},
  booktitle = {International Symposium on Physical Design},
  abbr = {ISPD},
  year = {2018},
  pdf = {/assets/pdf/ispd18.pdf}
}

2016

TCOMP

Design and applications for embedded networks-on-chip on FPGAs

Mohamed S. Abdelfattah, Andrew Bitar, Vaughn Betz

IEEE Transactions on Computers

@article{43,
  title = {Design and applications for embedded networks-on-chip on FPGAs},
  author = {Abdelfattah, Mohamed S. and Bitar, Andrew and Betz, Vaughn},
  journal = {IEEE Transactions on Computers},
  abbr = {TCOMP},
  year = {2016},
  pdf = {/assets/pdf/tcomp16.pdf}
}

FPL

LYNX: CAD for FPGA-based networks-on-chip

Mohamed S. Abdelfattah, Vaughn Betz

International Conference on Field-Programmable Logic and Applications

Nominated for the Michel Servit best paper award.

@inproceedings{45,
  title = {LYNX: CAD for FPGA-based networks-on-chip},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  booktitle = {International Conference on Field-Programmable Logic and Applications},
  abbr = {FPL},
  year = {2016},
  pdf = {/assets/pdf/fpl16.pdf},
  award = {Nominated for the Michel Servit best paper award.}
}

2015

FPT

Bringing programmability to the data plane: Packet processing with a NoC-enhanced FPGA

Andrew Bitar, Mohamed S. Abdelfattah, Vaughn Betz

International Conference on Field-Programmable Technology

@inproceedings{46,
  title = {Bringing programmability to the data plane: Packet processing with a NoC-enhanced FPGA},
  author = {Bitar, Andrew and Abdelfattah, Mohamed S. and Betz, Vaughn},
  booktitle = {International Conference on Field-Programmable Technology},
  abbr = {FPT},
  year = {2015},
  pdf = {/assets/pdf/fpt15.pdf}
}

FPL

Design and simulation tools for Embedded NOCs on FPGAs

Mohamed S. Abdelfattah, Andrew Bitar, Ange Yaghi, Vaughn Betz

International Conference on Field-Programmable Logic and Applications

@inproceedings{47,
  title = {Design and simulation tools for Embedded NOCs on FPGAs},
  author = {Abdelfattah, Mohamed S. and Bitar, Andrew and Yaghi, Ange and Betz, Vaughn},
  booktitle = {International Conference on Field-Programmable Logic and Applications},
  abbr = {FPL},
  year = {2015},
  pdf = {/assets/pdf/fpl15.pdf}
}

FPGA

Take the highway: Design for embedded NoCs on FPGAs

Mohamed S. Abdelfattah, Andrew Bitar, Vaughn Betz

International Symposium on Field-Programmable Gate Arrays

Best Paper Award

@inproceedings{49,
  title = {Take the highway: Design for embedded NoCs on FPGAs},
  author = {Abdelfattah, Mohamed S. and Bitar, Andrew and Betz, Vaughn},
  booktitle = {International Symposium on Field-Programmable Gate Arrays},
  abbr = {FPGA},
  year = {2015},
  pdf = {/assets/pdf/fpga15.pdf},
  award = {Best Paper Award}
}

TVLSI

Power analysis of embedded NoCs on FPGAs and comparison with custom buses

Mohamed S. Abdelfattah, Vaughn Betz

IEEE Transactions on Very Large-Scale Integration Systems

@article{50,
  title = {Power analysis of embedded NoCs on FPGAs and comparison with custom buses},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  journal = {IEEE Transactions on Very Large-Scale Integration Systems},
  abbr = {TVLSI},
  year = {2015},
  pdf = {/assets/pdf/tvlsi15.pdf}
}

2014

TRETS

Networks-on-Chip for FPGAs: Hard, Soft or Mixed?

Mohamed S. Abdelfattah, Vaughn Betz

ACM Transactions on Reconfigurable Technology and Systems

@article{52,
  title = {Networks-on-Chip for FPGAs: Hard, Soft or Mixed?},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  journal = {ACM Transactions on Reconfigurable Technology and Systems},
  abbr = {TRETS},
  year = {2014},
  pdf = {/assets/pdf/trets14.pdf}
}

IWOCL

Gzip on a chip: High performance lossless data compression on fpgas using opencl

Mohamed S. Abdelfattah, Andrei Hagiescu, Deshanand Singh

International Workshop on OpenCL

@inproceedings{53,
  title = {Gzip on a chip: High performance lossless data compression on fpgas using opencl},
  author = {Abdelfattah, Mohamed S. and Hagiescu, Andrei and Singh, Deshanand},
  booktitle = {International Workshop on OpenCL},
  abbr = {IWOCL},
  year = {2014},
  pdf = {/assets/pdf/iwocl14.pdf}
}

MICRO

The Case for Embedded Networks-on-Chip on FPGAs

Mohamed S. Abdelfattah, Vaughn Betz

IEEE Micro Magazine

@article{54,
  title = {The Case for Embedded Networks-on-Chip on FPGAs},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  journal = {IEEE Micro Magazine},
  abbr = {MICRO},
  year = {2014},
  pdf = {/assets/pdf/micro14.pdf}
}

2013

CARL

Augmenting FPGAs with Embedded Networks on Chip

Mohamed S. Abdelfattah, Vaughn Betz

Workshop on the Intersection of Computer Architecture and Reconfigurable Logic

@inproceedings{55,
  title = {Augmenting FPGAs with Embedded Networks on Chip},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  booktitle = {Workshop on the Intersection of Computer Architecture and Reconfigurable Logic},
  abbr = {CARL},
  year = {2013},
  pdf = {/assets/pdf/carl13.pdf}
}

FPL

The Power of Communication: Energy-Efficient NoCs for FPGAs

Mohamed S. Abdelfattah, Vaughn Betz

International Conference on Field-Programmable Logic and Applications

Stamatis Vassiliadis best paper award.
Selected as one of the most significant papers from the first 25 years of FPL.

@inproceedings{56,
  title = {The Power of Communication: Energy-Efficient NoCs for FPGAs},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  booktitle = {International Conference on Field-Programmable Logic and Applications},
  abbr = {FPL},
  year = {2013},
  pdf = {/assets/pdf/fpl13.pdf},
  award = {Stamatis Vassiliadis best paper award.},
  award2 = {Selected as one of the <a href=http://www.fpl2015.org/pdf/sig_papers.pdf>most significant papers</a> from the first 25 years of FPL.}
}

2012

FPT

Design tradeoffs for hard and soft FPGA-based Networks-on-Chip

Mohamed S. Abdelfattah, Vaughn Betz

International Conference on Field-Programmable Technology

@inproceedings{57,
  title = {Design tradeoffs for hard and soft FPGA-based Networks-on-Chip},
  author = {Abdelfattah, Mohamed S. and Betz, Vaughn},
  booktitle = {International Conference on Field-Programmable Technology},
  abbr = {FPT},
  year = {2012},
  pdf = {/assets/pdf/fpt12.pdf}
}

IOLTS

Transparent Structural Online Test for Reconfigurable Systems

Mohamed S. Abdelfattah, Lars Bauer, Claus Braun, Michael E Imhof, Michael A Kochte, Hongyan Zhang, Jörg Henkel, Hans-Joachim Wunderlich

International Symposium on On-Line Testing and Robust System Design

@inproceedings{58,
  title = {Transparent Structural Online Test for Reconfigurable Systems},
  author = {Abdelfattah, Mohamed S. and Bauer, Lars and Braun, Claus and Imhof, Michael E and Kochte, Michael A and Zhang, Hongyan and Henkel, Jörg and Wunderlich, Hans-Joachim},
  booktitle = {International Symposium on On-Line Testing and Robust System Design},
  abbr = {IOLTS},
  year = {2012},
  pdf = {/assets/pdf/iolts12.pdf}
}

2011

GeMiC

2.2 GHz LC VCO for RFID on a 0.5-μm digital gate-array designed for ultra-thin silicon substrates

Mohamed S. Abdelfattah, Damir Ferenci, Markus Grözing, Manfred Berroth, Cor Scherjon, Joachim Burghartz

German Microwave Conference

@inproceedings{59,
  title = {2.2 GHz LC VCO for RFID on a 0.5-μm digital gate-array designed for ultra-thin silicon substrates},
  author = {Abdelfattah, Mohamed S. and Ferenci, Damir and Grözing, Markus and Berroth, Manfred and Scherjon, Cor and Burghartz, Joachim},
  booktitle = {German Microwave Conference},
  abbr = {GeMiC},
  year = {2011},
  pdf = {/assets/pdf/gemic11.pdf}
}

2009

ProRISC

Design of a RF Transmitter for RFID Tags in a New Technology with Ultra Thin Silicon Substrates

Mohamed S. Abdelfattah, Damir Ferenci, Markus Grözing, Manfred Berroth, Cor Scherjon, Joachim N Burghartz

Workshop on Circuit Design and Digital Signal Processing

@inproceedings{61,
  title = {Design of a RF Transmitter for RFID Tags in a New Technology with Ultra Thin Silicon Substrates},
  author = {Abdelfattah, Mohamed S. and Ferenci, Damir and Grözing, Markus and Berroth, Manfred and Scherjon, Cor and Burghartz, Joachim N},
  booktitle = {Workshop on Circuit Design and Digital Signal Processing},
  abbr = {ProRISC},
  year = {2009},
  pdf = {/assets/pdf/prorisc09.pdf}
}

Books

Mohamed S. Abdelfattah and Vaughn Betz, "Embedded Networks-on-Chip for FPGAs", Chapter 6 in Reconfigurable Logic: Architecture Tools and Applications, P. Gaillardon, Taylor and Francis, 2015, pp. 149 - 184.

Theses

PhD Thesis: Embedded Networks on Chip for Field-Programmable Gate Arrays. 2016.
MSc Thesis: Evaluation of advanced techniques for structural FPGA self-test. 2011.
BSc Thesis: RF Transmitters for RFID Tags in a New Technology with Ultra Thin Silicon. 2009.

Patents

Systems and Methods for Low-Precision Neural Network Deployment. 2024. Application Number: US63718962
Apparatus and Method for Compute in BRAM for Multiply-Accumulate Operations. 2023. Application Number: US63454275
Method and Apparatus for Video Super Resolution. 2020. Application Number: UKX
Method and Apparatus for Neural Architecture Search. 2020. Application Number: UK2015231.0
Method and Apparatus for Analysing Neural Network Performance. 2020. Application Number: UK20199106.4
Method for Designing Accelerator Hardware. 2019. Application Number: UK1913353.7
Field Programmable Gate-Array with Embedded Network-on-Chip Hardware and Design Flow. 2015. Application Number: US14060253