8 citations · 12 across the 5 of their papers we have counts for
7 papers
VisionMX: Unlocking Microscaling Post-Training Quantization for Vision Models
Elad Dror Cohen, Ofir Gordon, Lior Dikstein +2
Microscaling (MX) formats are emerging as a hardware-supported approach to efficient training and inference. They combine low-precision elements with shared block scales, but their…
Adaptive Latent Capacity for World Models
Idan Achituve, Lior Dikstein, Idit Diamant +2
We introduce Adaptive LeWorldModel (ALeWM), a world model based on a joint-embedding predictive architecture (JEPA) that learns to concentrate predictive information in compact pre…
LATMiX: Learnable Affine Transformations for Microscaling Quantization of LLMs
Ofir Gordon, Lior Dikstein, Arnon Netzer +2
Post-training quantization (PTQ) is a widely used approach for reducing the memory and compute costs of large language models (LLMs). Recent studies have shown that applying invert…
Data Generation for Hardware-Friendly Post-Training Quantization
Lior Dikstein, Ariel Lapid, Arnon Netzer +1
Zero-shot quantization (ZSQ) using synthetic data is a key approach for post-training quantization (PTQ) under privacy and security constraints. However, existing data generation m…
HPTQ: Hardware-Friendly Post Training Quantization
Hai Victor Habi, Reuven Peretz, Elad Cohen +5
Neural network quantization enables the deployment of models on edge devices. An essential requirement for their hardware efficiency is that the quantizers are hardware-friendly: u…
On State Dependent Broadcast Channels with Cooperation
Lior Dikstein, Haim H. Permuter, Yossef Steinberg
In this paper, we investigate problems of communication over physically degraded, state-dependent broadcast channels (BCs) with cooperating decoders. Two different setups are consi…
MAC with Action-Dependent State Information at One Encoder
Lior Dikstein, Haim H. Permuter, Shlomo Shamai
Problems dealing with the ability to take an action that affects the states of state-dependent communication channels are of timely interest and importance. Therefore, we extend th…