Aditya Tomar

I'm a 4th year undergraduate at UC Berkeley studying EECS.

My research broadly lies in efficient deep learning. Currently, I'm working at BAIR under Prof. Kurt Keutzer and Dr. Amir Gholami. Previously, I had the pleasure of working under Prof. Abhinav Bhatele at PSSG on memory-efficient adaptive optimization algorithms and efficient parallel training methods for LLMs.

I am currently a research intern at NVIDIA in the Applied Deep Learning Research team in Santa Clara. I'm working on large-scale efficient MoE inference for Blackwell NVL72 class systems. Please feel free to reach out if you'd like to chat.

* = equal contribution

XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization

Aditya Tomar*, Coleman Hooper*, Minjae Lee, Haocheng Xi, Rishabh Tiwari, Wonjun Kang, Luca Manolache, Michael W. Mahoney, Kurt Keutzer, Amir Gholami

Under Review

PDF X

Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models

Rishabh Tiwari*, Aditya Tomar*, Udbhav Bamba*, Monishwaran Maheswaran, Heng Yang, Michael W. Mahoney, Kurt Keutzer, Amir Gholami

ICML 2026

PDF

QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache

Rishabh Tiwari*, Haocheng Xi*, Aditya Tomar*, Coleman Hooper, Sehoon Kim, Maxwell Horton, Mahyar Najibi, Michael W. Mahoney, Kurt Keutzer, Amir Gholami

ICML 2025

PDF

Eve: Less Memory, Same Might

Aditya Tomar, Siddharth Singh, Tom Goldstein, Abhinav Bhatele

(ACM SRC Poster) SC 2024

PDF

Democratizing AI: Open-source Scalable LLM Training on GPU-based Supercomputers

Siddharth Singh, Prajwal Singhania, Aditya Ranjan, John Kirchenbauer, Jonas Geiping, Yuxin Wen, Neel Jain, Abhimanyu Hans, Manli Shu, Aditya Tomar, Tom Goldstein, Abhinav Bhatele

(ACM Gordon Bell Prize Finalist) SC 2024

PDF

Residual Context Diffusion Language Models

Yuezhou Hu, Harman Singh, Monishwaran Maheswaran, Haocheng Xi, Coleman Hooper, Jintao Zhang, Aditya Tomar, Michael W. Mahoney, Sewon Min, Mehrdad Farajtabar, Kurt Keutzer, Amir Gholami, Chenfeng Xu

ICML 2026

PDF

LoSA: Locality Aware Sparse Attention for Block-Wise Diffusion Language Models

Haocheng Xi, Harman Singh, Yuezhou Hu, Coleman Hooper, Rishabh Tiwari, Aditya Tomar, Minjae Lee, Wonjun Kang, Michael W. Mahoney, Chenfeng Xu, Kurt Keutzer, Amir Gholami

ICML 2026

PDF

Beyond Next-Token Prediction: A Performance Characterization of Diffusion versus Autoregressive Language Models

Minseo Kim, Coleman Hooper, Aditya Tomar, Chenfeng Xu, Mehrdad Farajtabar, Michael W. Mahoney, Kurt Keutzer, Amir Gholami

Preprint

PDF YT

Automated Programmatic Performance Analysis of Parallel Programs

Onur Cankur, Aditya Tomar, Daniel Nichols, Connor Scully-Allison, Katherine E. Isaacs, Abhinav Bhatele

Under Review

PDF