iiinova — where your digital self keeps evolving

iiinova is part of the Atmosphere — an open, public network where you own your identity.

Samin Aref

@samin.mas.to.ap.brid.gy

Researcher and educator Faculty at the University of Toronto Networks, Machine Learning, Operations Research, and (Social) Data Science (Reblog⇏Endorse) [bridged from mas.to/@samin on the fediverse by fed.brid.gy ]

Posts Replies Media

Samin Aref @samin.mas.to.ap.brid.gy · 11/08/2025

link.springer.com/chapter/10.1007/9…

link.springer.com

Enhancing Ultra-Low-Bit Quantization of Large Language Models Through Saliency-Aware Partial Retraining

The growing use of large language models has raised environmental and economic concerns about their intensity of resource usage during inference. Serving these models to each user requires substantial energy and water for cooling. Model compression techniques like...

000 0

Samin Aref @samin.mas.to.ap.brid.gy · 24/04/2025

make them more resource efficient at the cost of potential performance reduction. In a new preprint, we extended an existing frontier method (ApiQ) for quantizing weights of LLMs down to 1 or 2 bits. On LLaMA-2 models, our method covers 7.5-10.8% of the accuracy gap between the full-precision […]

mas.to

Original post on mas.to

100 1

Samin Aref @samin.mas.to.ap.brid.gy · 24/04/2025

🧵 The growing use of Large Language Models (LLMs) has raised economic and environmental concerns about their resource inefficiency in inference. Even after training, serving LLMs to each user among millions of users requires substantial energy and water for cooling to […] [Original post on mas.to]

100 1

Samin Aref @samin.mas.to.ap.brid.gy · 20/11/2024

In a joint work with SickKids, we investigated the organization of chromosomes within human cells using large-scale genomic data. We analyzed 3.8 million chromosomal interactions across 62 Hi-C datasets using supervised learning, network clustering, and computational […] [Original post on mas.to]

010 1

Samin Aref @samin.mas.to.ap.brid.gy · 20/11/2024

In a joint work with The Hospital for Sick Children, we investigated the organization of chromosomes within human cells using large-scale genomic data. We analyzed 3.8 million chromosomal interactions across 62 Hi-C datasets using supervised learning, network clustering, and computational […]

mas.to

Original post on mas.to

000 0