Ali Abbasi

Vanderbilt University, Department of Computer Science.

Hi, I’m Ali. I’m a computer science Ph.D. candidate at Vanderbilt University, where I work with Prof. Soheil Kolouri in the MINT Lab. Before that I picked up an M.S. at Vanderbilt along the way (2024) and a B.S. in electrical engineering from the University of Tehran (2020). I also spent the summer of 2024 as an intern at Microsoft Research in Redmond, WA.

My research is focused on methods that make large models more efficient to store and run, both at scale in the data center and on resource-constrained edge devices. My recent work covers low-rank factorization of LLMs (ZS-SVD, IO-SVD), activation quantization (ConQuR), and expert selection for mixture-of-experts serving (BASE). My earlier work includes compact reparameterizations of neural networks (PRANC, MCNC) and dataset distillation.

As for hobbies, I enjoy running and lifting weights, though I would cancel all my plans if someone invites me for a poker night.

news

Sep 2026	New preprint: BASE, batch-aware expert selection for efficient MoE decoding.
Sep 2026	ConQuR accepted to NeurIPS 2026.
Jun 2026	Two papers at the SynData4CV workshop at CVPR 2026: One Category One Prompt and Diffusion-Augmented Coreset Expansion.
May 2026	New preprint: IO-SVD, input-output whitened SVD for adaptive-rank LLM compression.
Apr 2026	Zero Sum SVD accepted to ICML 2026.
Mar 2026	New preprint: Vector-Quantized Soft Label Compression for Dataset Distillation.
Jan 2025	MCNC accepted to ICLR 2025.
Dec 2024	New preprint: Diffusion-Augmented Coreset Expansion, from my internship at Microsoft.
Aug 2024	Received my M.S. in computer science from Vanderbilt University, en route to the Ph.D.
Jun 2024	Started a research internship at Microsoft in Redmond, WA.
Mar 2024	New preprint: One Category One Prompt, dataset distillation using diffusion models.
Feb 2024	BrainWash accepted to CVPR 2024.
Jul 2023	PRANC accepted to ICCV 2023.
Dec 2021	I am delighted to announce that beginning the January of 2022, I will be joining the wonderful scholarly community of Vanderbilt University as a Ph.D. student in the department of computer science! 😃

selected publications

conference papers

NeurIPS

ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs

Thrash, Chayne, Abbasi, Ali, and Kolouri, Soheil

In Advances in Neural Information Processing Systems (NeurIPS) 2026

Abs arXiv

LLM activation quantization learning optimized rotations that align activations with hypercube corners; 19.2× faster quantization than SpinQuant with lower perplexity across Llama families.
ICML

Zero Sum SVD: Balancing Loss Sensitivity for Low Rank LLM Compression

Abbasi, Ali, Thrash, Chayne, Qin, Haoran, Sharma, Shansita, Seifi, Sepehr, and Kolouri, Soheil

In International Conference on Machine Learning (ICML) 2026

Abs arXiv Website

Selects singular components across all layers with a zero-sum rule on first-order loss estimates; at 40% retention, 5.86× the uncompressed throughput and 16% lower WikiText-2 perplexity than SVD-LLM.

On-device demo: NVIDIA Jetson edge device

LLaMA-2-7B-Chat compressed with Zero-Sum SVD, keeping about 50% of the parameters in low-rank form with a different rank for each weight matrix and 8-bit quantization, running locally on an NVIDIA Jetson edge device.

Compared with the uncompressed model:

1.30×peak speedupat 1024 input tokens

20–23%latency reductionend to end

1.31×faster per tokentime per output token
ICLR

MCNC: Manifold-Constrained Reparameterization for Neural Compression

Thrash, Chayne, Abbasi, Ali, Andreas, Reed, Nooralinejad, Parsa, Koohpayegani, Soroush Abbasi, Pirsiavash, Hamed, and Kolouri, Soheil

In International Conference on Learning Representations (ICLR) 2025

Abs arXiv PDF

Constrains parameters to a low-dimensional nonlinear manifold generated by a frozen random network; +22.3 accuracy points over PLATON pruning at 50× compression (ViT-Ti, ImageNet-100).
CVPR

BrainWash: A Poisoning Attack to Forget in Continual Learning

Abbasi, Ali, Nooralinejad, Parsa, Pirsiavash, Hamed, and Kolouri, Soheil

In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024

Abs arXiv PDF

Untargeted poisoning attack forcing forgetting in continual learners via model inversion of past data; up to 80% more forgetting than uniform noise.
ICCV

PRANC: Pseudo Random Networks for Compacting Deep Models

Nooralinejad, Parsa, Abbasi, Ali, Koohpayegani, Soroush Abbasi, Pourahmadi Meibodi, Kossar, Khan, Rana Muhammad Shahroz, Kolouri, Soheil, and Pirsiavash, Hamed

In IEEE/CVF International Conference on Computer Vision (ICCV) 2023

Abs arXiv PDF

Reparameterizes weights as a linear mixture of frozen random networks regenerated from a seed; +8.1 accuracy points over HashedNet at about 100× compression (ResNet-18, ImageNet-100).

preprints

arXiv

BASE: Batch-Aware Selection of Experts Using Predicted Removal Error for Efficient MoE Decoding

Abbasi, Ali, Shi, Justin, and Kolouri, Soheil

arXiv preprint arXiv:2609.36222 2026

Abs arXiv

Batch-aware MoE expert selection from predicted removal error, with custom GPU kernels; +29.5 accuracy points over the strongest baseline, or 60% faster than dense within 0.4 points (Qwen3-30B-A3B).
arXiv

IO-SVD: Input-Output Whitened SVD for Adaptive-Rank LLM Compression

Abbasi, Ali, Thrash, Chayne, Qin, Haoran, Pirsiavash, Hamed, and Kolouri, Soheil

arXiv preprint arXiv:2605.15626 2026

Abs arXiv

LLM/VLM compression via KL-aware input-output whitening and adaptive ranks; +10.2 avg. accuracy points over WSVD (SmolVLM-2B); 4.34× decode speedup with KV-cache compression (LLaMA-2-7B).

workshop papers

CVPRW

One Category One Prompt: Dataset Distillation using Diffusion Models

Abbasi, Ali, Shahbazi, Ashkan, Pirsiavash, Hamed, and Kolouri, Soheil

In CVPR Workshops (SynData4CV) 2026

Abs arXiv

Distills each class into a learnable text prompt for a diffusion model; +12.6 accuracy points over RDED on ImageNet-1K at 50 images/class (ResNet-18, stored soft labels).
CVPRW

Diffusion-Augmented Coreset Expansion for Scalable Dataset Distillation

Abbasi, Ali, Imani, Shima, An, Chenyang, Mahalingam, Gayathri, Shrivastava, Harsh, Diesendruck, Maurice, Pirsiavash, Hamed, Sharma, Pramod, and Kolouri, Soheil

In CVPR Workshops (SynData4CV) 2026

Abs arXiv

Expands compact low-resolution coresets with uncertainty-guided diffusion augmentation; +10.1 accuracy points over RDED on ImageNet-1K at 10 images/class (ResNet-18).