Arash Akbari

Palace of Fine Arts, San Francisco

I am a PhD student in Computer Engineering at Northeastern University, advised by Prof. Yanzhi Wang. My research sits at the intersection of efficient deep learning and physical AI, with a focus on accelerating vision-language-action (VLA) models and world action models and making their edge deployment more energy-efficient.

I earned my M.S. in Computer Engineering from Northeastern University and my B.Sc. in Computer Science from the University of Tehran. I have also been a research intern at CNRS with Timothée Masquelier and at CISPA with Xiao Zhang.

Feel free to reach out via email if you’d like to discuss research or potential collaborations.

News

Sep 24, 2026	ActQuant and Palette got accepted in NeurIPS 2026! See you in Atlanta!
Aug 26, 2026	Joined Honda Research Institute USA, Inc. as a Research Intern.
May 21, 2026	Our paper Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs was accepted at GLSVLSI 2026.
Apr 01, 2026	Earned my M.S. in Computer Engineering from Northeastern University, with a concentration in Computer Vision, Machine Learning, and Algorithms (CVLA), along the way to my Ph.D.
Mar 15, 2025	Our paper Moxin-LLM was accepted to SCI-FM @ ICLR’25.
Sep 01, 2024	Started my direct PhD in Computer Engineering at Northeastern University with Prof. Yanzhi Wang.

Selected Publications see all publications

Preprint

Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection

Arash Akbari, Arman Akbari, Jingwu Luo, Yuhao Lei, Yi Gao, Weiwei Chen, Xuan Zhang, Zhenman Fang, Geng Yuan, and Yanzhi Wang

arXiv preprint arXiv:2609.33269, 2026

PDF
Preprint

Flash-WAM: Modality-Aware Distillation for World Action Models

Arman Akbari, Ci Zhang, Arash Akbari, Lin Zhao, Yixiao Chen, Weiwei Chen, Xuan Zhang, Geng Yuan, and Yanzhi Wang

arXiv preprint arXiv:2606.05254, 2026

PDF Website
Preprint

PhyGround: Benchmarking Physical Reasoning in Generative World Models

Juyi Lin^*, Arash Akbari^*, Yumei He, Lin Zhao, Haichao Zhang, Arman Akbari, Xingchen Xu, Zoe Y. Lu, Enfu Nan, Hokin Deng, Edmund Yeh, Sarah Ostadabbas, Yun Fu, Jennifer Dy, Pu Zhao, and Yanzhi Wang

arXiv preprint arXiv:2605.10806, 2026

PDF Code Website
NeurIPS

ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models

Arash Akbari, Arman Akbari, Masih Eskandar, Qitao Tan, Yixiao Chen, Jingwu Luo, Bertha Pangaribuan, Liyun Zhang, Jennifer Dy, Geng Yuan, Xue Lin, Gaowen Liu, Stratis Ioannidis, and Yanzhi Wang

In Conference on Neural Information Processing Systems (NeurIPS), 2026

PDF Website
NeurIPS

Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs

Qitao Tan, Xiaoying Song, Arman Akbari, Arash Akbari, Yanzhi Wang, Xiaoming Zhai, Lingzi Hong, Zhen Xiang, Jin Lu, and Geng Yuan

In Conference on Neural Information Processing Systems (NeurIPS), 2026

PDF Code Demo
GLSVLSI

Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs

Amir Taherin, Juyi Lin, Arash Akbari, Arman Akbari, Pu Zhao, Weiwei Chen, David Kaeli, and Yanzhi Wang

In ACM Great Lakes Symposium on VLSI (GLSVLSI), 2026

PDF
ICANN

VOTE: Vision-Language-Action Optimization with Trajectory Ensemble Voting

Juyi Lin, Amir Taherin, Arash Akbari, Arman Akbari, Lei Lu, Guangyu Chen, Taskin Padir, Xiaomeng Yang, Weiwei Chen, Yiqian Li, Xue Lin, David Kaeli, Pu Zhao, and Yanzhi Wang

In International Conference on Artificial Neural Networks (ICANN), 2026

PDF Code
ICLR

7B Fully Open Source Moxin-LLM – From Pretraining to GRPO-based Reinforcement Learning Enhancement

Pu Zhao, Xuan Shen, Zhenglun Kong, Yixin Shen, Sung-En Chang, Arash Akbari, Timothy Rupprecht, Lei Lu, Enfu Nan, Changdi Yang, and others

In SCI-FM @ ICLR Workshop, 2025

PDF Code