Ashish Sirasao

Researcher, AIG, Amd inc

Joined

March 2023

Names

Ashish Sirasao (Preferred)

Emails

****@amd.com (Confirmed)

Personal Links

DBLP

Career & Education History

Researcher

AIG, Amd inc (amd.com)

2022 – Present

MS student

Indian Institute of Technology, Bombay (iitb.ac.in)

1991 – 1993

Advisors, Relations & Conflicts

Coworker

Paolo d alberto

1998 – 2022

Expertise

Efficient Deep learning hardware

2016 – Present

Compilers for custom accelerators

2016 – Present

Publications

Exploring the Dynamic Scheduling Space of Real-Time Generative AI Applications on Emerging Heterogeneous Systems
Hyukjun Kwon, Rajeev Patwari, Rachid Karami, Ashish Sirasao
- OpenReview Archive Direct Upload
- Readers: Everyone
Forecasting LLM Inference Performance via Hardware-Agnostic Analytical Modeling
Rajeev Patwari, Devleena Das, Ashish Sirasao
- OpenReview Archive Direct Upload
- Readers: Everyone
Recover-LoRA: Data-Free Accuracy Recovery of Degraded Language Models via Low-Rank Adaptation
Rajeev Patwari, Devleena Das, Ashish Sirasao
- EMNLP 2025
- Readers: Everyone
KV Pareto: Systems-Level Optimization of KV Cache and Model Compression for Long Context Inference
Rajeev Patwari, Devleena Das, Sai Gokhale, Elliott Delaye, Ashish Sirasao
- EACL 2026
- Readers: Everyone
Dual LoRA: Enhancing LoRA with Magnitude and Direction Updates
Yixing Xu, Chao Li, Xuanwu Yin, Spandan Tiwari, Dong Li, Ashish Sirasao, Emad Barsoum
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
DiP-GO: A Diffusion Pruner via Few-step Gradient Optimization
haoweiz, Dehua Tang, Ji Liu, Mingjie Lu, Jintu Zheng, Jinzhang Peng, Dong Li, Yu Wang, Fan Jiang, Lu Tian, Spandan Tiwari, Ashish Sirasao, Jun-Hai Yong, Bin Wang, Emad Barsoum
- NeurIPS 2024 poster
- Readers: Everyone
Customizing Text-to-Image Generation with Inverted Interaction
mengmeng Ge, Xu Jia, Takashi Isobe, Xiaomin Li, Qinghe Wang, Jing Mu, Dong Zhou, liwang Amd, Huchuan Lu, Lu Tian, Ashish Sirasao, Emad Barsoum
- MM2024 Poster
- Readers: Everyone
Enhancing Vision Transformer: Amplifying Non-Linearity in Feedforward Network Module
Yixing Xu, Chao Li, Dong Li, Xiao Sheng, Fan Jiang, Lu Tian, Ashish Sirasao, Emad Barsoum
- ICML 2024 Poster
- Readers: Everyone
Enhancing One-Shot Pruned Generative Pre-training Language Models through Sparse-Dense-Sparse Mechanism
Guanchen Li, Xiandong Zhao, Zeping Li, Dong Li, Lu Tian, Jie He, Ashish Sirasao
- Submitted to ICLR 2024
- Readers: Everyone
Improving MLP Module in Vision Transformer
Yixing Xu, Chao Li, Dong Li, Xiao Sheng, Fan Jiang, Lu Tian, Ashish Sirasao
- Submitted to ICLR 2024
- Readers: Everyone

Co-Authors

View all 37 co-authors