Saleh Ashkboos

37 papers A* 8A 4Journal 23Unranked 2

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Beyond Outliers: A Study of Optimizers Under Quantization. CoRR Georgios Vlassis, Saleh Ashkboos, Alexandra Volkova, Torsten Hoefler, Dan Alistarh
2025	J	`jnl`	Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantization. CoRR Vage Egiazarian, Roberto L. Castro, Denis Kuznedelev, Andrei Panferov, Eldar Kurtic, Shubhra Pandit, Alexandre Noll Marques, Mark Kurtz, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh
2025	A	`conf`	Energy-Optimal and Low-Depth Algorithmic Primitives for Spatial Dataflow Architectures. IPDPS Lukas Gianinazzi, Tal Ben-Nun, Maciej Besta, Saleh Ashkboos, Yves Baumann, Piotr Luczynski, Torsten Hoefler
2025	J	`jnl`	HALO: Hadamard-Assisted Lower-Precision Optimization for LLMs. CoRR Saleh Ashkboos, Mahdi Nikdan, Soroush Tabesh, Roberto L. Castro, Torsten Hoefler, Dan Alistarh
2025	J	`jnl`	Quartet: Native FP4 Training Can Be Optimal for Large Language Models. CoRR Roberto L. Castro, Andrei Panferov, Soroush Tabesh, Oliver Sieberling, Jiale Chen, Mahdi Nikdan, Saleh Ashkboos, Dan Alistarh
2024	J	`jnl`	Arrow Matrix Decomposition: A Novel Approach for Communication-Efficient Sparse Matrix Multiplication. CoRR Lukas Gianinazzi, Alexandros Nikolaos Ziogas, Langwen Huang, Piotr Luczynski, Saleh Ashkboos, Florian Scheidl, Armon Carigiet, Chio Ge, Nabil Abubaker, Maciej Besta, Tal Ben-Nun, Torsten Hoefler
2024	—	`conf`	Computational Bottlenecks of Training Small-scale Large Language Models. ENLSP Saleh Ashkboos, Seyed-Iman Mirzadeh, Keivan Alizadeh-Vahid, Mohammad Hossein Sekhavat, Moin Nabi, Mehrdad Farajtabar, Fartash Faghri
2024	J	`jnl`	Computational Bottlenecks of Training Small-scale Large Language Models. CoRR Saleh Ashkboos, Iman Mirzadeh, Keivan Alizadeh, Mohammad Hossein Sekhavat, Moin Nabi, Mehrdad Farajtabar, Fartash Faghri
2024	J	`jnl`	EfQAT: An Efficient Framework for Quantization-Aware Training. CoRR Saleh Ashkboos, Bram Verhoef, Torsten Hoefler, Evangelos Eleftheriou, Martino Dazzi
2024	A*	`conf`	QUIK: Towards End-to-end 4-Bit Inference on Generative Large Language Models. EMNLP Saleh Ashkboos, Ilia Markov, Elias Frantar, Tingxuan Zhong, Xincheng Wang, Jie Ren, Torsten Hoefler, Dan Alistarh
2024	A*	`conf`	QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs. NeurIPS Saleh Ashkboos, Amirkeivan Mohtashami, Maximilian L. Croci, Bo Li, Pashmina Cameron, Martin Jaggi, Dan Alistarh, Torsten Hoefler, James Hensman
2024	J	`jnl`	QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs. CoRR Saleh Ashkboos, Amirkeivan Mohtashami, Maximilian L. Croci, Bo Li, Martin Jaggi, Dan Alistarh, Torsten Hoefler, James Hensman
2024	A*	`conf`	SliceGPT: Compress Large Language Models by Deleting Rows and Columns. ICLR Saleh Ashkboos, Maximilian L. Croci, Marcelo Gennari Do Nascimento, Torsten Hoefler, James Hensman
2024	J	`jnl`	SliceGPT: Compress Large Language Models by Deleting Rows and Columns. CoRR Saleh Ashkboos, Maximilian L. Croci, Marcelo Gennari Do Nascimento, Torsten Hoefler, James Hensman
2024	A*	`conf`	SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression. ICLR Tim Dettmers, Ruslan Svirschevski, Vage Egiazarian, Denis Kuznedelev, Elias Frantar, Saleh Ashkboos, Alexander Borzunov, Torsten Hoefler, Dan Alistarh
2023	A*	`conf`	OPTQ: Accurate Quantization for Generative Pre-trained Transformers. ICLR Elias Frantar, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh
2023	J	`jnl`	STen: Productive and Efficient Sparsity in PyTorch. CoRR Andrei Ivanov, Nikoli Dryden, Tal Ben-Nun, Saleh Ashkboos, Torsten Hoefler
2023	J	`jnl`	SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression. CoRR Tim Dettmers, Ruslan Svirschevski, Vage Egiazarian, Denis Kuznedelev, Elias Frantar, Saleh Ashkboos, Alexander Borzunov, Torsten Hoefler, Dan Alistarh
2023	J	`jnl`	Towards End-to-end 4-Bit Inference on Generative Large Language Models. CoRR Saleh Ashkboos, Ilia Markov, Elias Frantar, Tingxuan Zhong, Xincheng Wang, Jie Ren, Torsten Hoefler, Dan Alistarh
2022	J	`jnl`	ENS-10: A Dataset For Post-Processing Ensemble Weather Forecast. CoRR Saleh Ashkboos, Langwen Huang, Nikoli Dryden, Tal Ben-Nun, Peter Dueben, Lukas Gianinazzi, Luca Kummer, Torsten Hoefler
2022	A*	`conf`	ENS-10: A Dataset For Post-Processing Ensemble Weather Forecasts. NeurIPS Saleh Ashkboos, Langwen Huang, Nikoli Dryden, Tal Ben-Nun, Peter Dueben, Lukas Gianinazzi, Luca Kummer, Torsten Hoefler
2022	J	`jnl`	GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers. CoRR Elias Frantar, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh
2022	A*	`conf`	Motif Prediction with Graph Neural Networks. KDD Maciej Besta, Raphael Grob, Cesare Miglioli, Nicola Bernold, Grzegorz Kwasniewski, Gabriel Gjini, Raghavendra Kanakagiri, Saleh Ashkboos, Lukas Gianinazzi, Nikoli Dryden, Torsten Hoefler
2022	A	`conf`	ProbGraph: High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations. SC Maciej Besta, Cesare Miglioli, Paolo Sylos Labini, Jakub Tetek, Patrick Iff, Raghavendra Kanakagiri, Saleh Ashkboos, Kacper Janda, Michal Podstawski, Grzegorz Kwasniewski, Niels Gleinig, Flavio Vella, Onur Mutlu, Torsten Hoefler
2022	J	`jnl`	ProbGraph: High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations. CoRR Maciej Besta, Cesare Miglioli, Paolo Sylos Labini, Jakub Tetek, Patrick Iff, Raghavendra Kanakagiri, Saleh Ashkboos, Kacper Janda, Michal Podstawski, Grzegorz Kwasniewski, Niels Gleinig, Flavio Vella, Onur Mutlu, Torsten Hoefler
2022	J	`jnl`	The spatial computer: A model for energy-efficient parallel computation. CoRR Lukas Gianinazzi, Tal Ben-Nun, Saleh Ashkboos, Yves Baumann, Piotr Luczynski, Torsten Hoefler
2021	—	`conf`	Degree-based Feature Is All You Need: Science4Cast Report. IEEE BigData Milad Aghajohari, Mohammad Sadegh Akhondzadeh, Saleh Ashkboos, Kamran Chitsaz
2021	J	`jnl`	Flare: Flexible In-Network Allreduce. CoRR Daniele De Sensi, Salvatore Di Girolamo, Saleh Ashkboos, Shigang Li, Torsten Hoefler
2021	A	`conf`	Flare: flexible in-network allreduce. SC Daniele De Sensi, Salvatore Di Girolamo, Saleh Ashkboos, Shigang Li, Torsten Hoefler
2021	J	`jnl`	Motif Prediction with Graph Neural Networks. CoRR Maciej Besta, Raphael Grob, Cesare Miglioli, Nicola Bernold, Grzegorz Kwasniewski, Gabriel Gjini, Raghavendra Kanakagiri, Saleh Ashkboos, Lukas Gianinazzi, Nikoli Dryden, Torsten Hoefler
2021	J	`jnl`	Multi-way sparsest cut problem on trees with a control on the number of parts and outliers. Discret. Appl. Math. Ramin Javadi, Saleh Ashkboos
2021	A*	`conf`	New Bounds For Distributed Mean Estimation and Variance Reduction. ICLR Peter Davies, Vijaykrishna Gurunanthan, Niusha Moshrefi, Saleh Ashkboos, Dan Alistarh
2020	J	`jnl`	Distributed Mean Estimation with Optimal Error Bounds. CoRR Dan Alistarh, Saleh Ashkboos, Peter Davies
2019	A	`conf`	SparCML: high-performance sparse communication for machine learning. SC Cédric Renggli, Saleh Ashkboos, Mehdi Aghagolzadeh, Dan Alistarh, Torsten Hoefler
2017	J	`jnl`	An Efficient Parallel Data Clustering Algorithm Using Isoperimetric Number of Trees. CoRR Ramin Javadi, Saleh Ashkboos
2017	J	`jnl`	Minimum Cuts of Distance-Regular Digraphs. Electron. J. Comb. Saleh Ashkboos, Gholam Reza Omidi, Fateme Shafiei, Khosro Tajbakhsh
2017	J	`jnl`	Multi-way sparsest cut problem on trees with a control on the number of parts and outliers. CoRR Ramin Javadi, Saleh Ashkboos