Souvik Kundu - RankMe

50 papers A* 10A 1Misc 1Journal 33Unranked 5

Year	Rank	Type	Title / Venue / Authors
2025	A	`conf`	Accelerating LLM Inference with Flexible N:M Sparsity via A Fully Digital Compute-in-Memory Accelerator. ISLPED Akshat Ramachandran, Souvik Kundu, Arnab Raha, Shamik Kundu, Deepak K. Mathaikutty, Tushar Krishna
2025	J	`jnl`	Accelerating LLM Inference with Flexible N:M Sparsity via A Fully Digital Compute-in-Memory Accelerator. CoRR Akshat Ramachandran, Souvik Kundu, Arnab Raha, Shamik Kundu, Deepak A. Mathaikutty, Tushar Krishna
2025	J	`jnl`	Assortment of Attention Heads: Accelerating Federated PEFT with Head Pruning and Strategic Client Selection. CoRR Yeshwanth Venkatesha, Souvik Kundu, Priyadarshini Panda
2025	J	`jnl`	Assortment of Attention Heads: Accelerating Federated PEFT with Head Pruning and Strategic Client Selection. Trans. Mach. Learn. Res. Yeshwanth Venkatesha, Souvik Kundu, Priyadarshini Panda
2025	J	`jnl`	AttentionBreaker: Adaptive Evolutionary Optimization for Unmasking Vulnerabilities in LLMs through Bit-Flip Attacks. Trans. Mach. Learn. Res. Sanjay Das, Swastik Bhattacharya, Souvik Kundu, Shamik Kundu, Anand Menon, Arnab Raha, Kanad Basu
2025	J	`jnl`	CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing. CoRR Wenhao Zheng, Yixiao Chen, Weitong Zhang, Souvik Kundu, Yun Li, Zhengzhong Liu, Eric P. Xing, Hongyi Wang, Huaxiu Yao
2025	J	`jnl`	COBRA: Catastrophic Bit-flip Reliability Analysis of State-Space Models. CoRR Sanjay Das, Swastik Bhattacharya, Shamik Kundu, Arnab Raha, Souvik Kundu, Kanad Basu
2025	J	`jnl`	Enhancing Large Language Models for Hardware Verification: A Novel SystemVerilog Assertion Dataset. CoRR Anand Menon, Samit S. Miftah, Shamik Kundu, Souvik Kundu, Amisha Srivastava, Arnab Raha, Gabriel Theodor Sonnenschein, Suvadeep Banerjee, Deepak Mathaikutty, Kanad Basu
2025	J	`jnl`	Fast and Cost-effective Speculative Edge-Cloud Decoding with Early Exits. CoRR Yeshwanth Venkatesha, Souvik Kundu, Priyadarshini Panda
2025	J	`jnl`	Fast and Cost-effective Speculative Edge-Cloud Decoding with Early Exits. Trans. Mach. Learn. Res. Yeshwanth Venkatesha, Souvik Kundu, Priyadarshini Panda
2025	J	`jnl`	Improving the Throughput of Diffusion-based Large Language Models via a Training-Free Confidence-Aware Calibration. CoRR Jucheng Shen, Gaurav Sarkar, Yeonju Ro, Sharath Nittur Sridhar, Zhangyang Wang, Aditya Akella, Souvik Kundu
2025	—	`conf`	LAMB: A Training-Free Method to Enhance the Long-Context Understanding of SSMs via Attention-Guided Token Filtering. ACL (2) Zhifan Ye, Zheng Wang, Kejing Xia, Jihoon Hong, Leshu Li, Lexington Allen Whalen, Cheng Wan, Yonggan Fu, Yingyan Celine Lin, Souvik Kundu
2025	J	`jnl`	LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models. CoRR Sihwan Park, Doohyuk Jang, Sungyub Kim, Souvik Kundu, Eunho Yang
2025	A*	`conf`	LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding. ICLR Doohyuk Jang, Sihwan Park, June Yong Yang, Yeonsung Jung, Jihun Yun, Souvik Kundu, Sungyub Kim, Eunho Yang
2025	—	`conf`	LLM-NPU: Towards Efficient Foundation Model Inference on Low-Power Neural Processing Units. COINS Arnab Raha, Souvik Kundu, Sharath Nittur Sridhar, Shamik Kundu, Soumendu Kumar Ghosh, Alessandro Palla, Arghadip Das, Darren Crews, Deepak A. Mathaikutty
2025	A*	`conf`	MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization. ISCA Akshat Ramachandran, Souvik Kundu, Tushar Krishna
2025	J	`jnl`	On Evaluating Performance of LLM Inference Serving Systems. CoRR Amey Agrawal, Nitin Kedia, Anmol Agarwal, Jayashree Mohan, Nipun Kwatra, Souvik Kundu, Ramachandran Ramjee, Alexey Tumanov
2025	A*	`conf`	On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention for Long-Context LLM Serving. ICML Yeonju Ro, Zhenyu Zhang, Souvik Kundu, Zhangyang Wang, Aditya Akella
2025	J	`jnl`	On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention. CoRR Yeonju Ro, Zhenyu Zhang, Souvik Kundu, Zhangyang Wang, Aditya Akella
2025	J	`jnl`	OuroMamba: A Data-Free Quantization Framework for Vision Mamba Models. CoRR Akshat Ramachandran, Mingyu Lee, Huan Xu, Souvik Kundu, Tushar Krishna
2025	J	`jnl`	PRISM: Enhancing Protein Inverse Folding through Fine-Grained Retrieval on Structure-Sequence Multimodal Representations. CoRR Sazan Mahbub, Souvik Kundu, Eric P. Xing
2025	J	`jnl`	Pushing the Limits of Sparsity: A Bag of Tricks for Extreme Pruning. Trans. Mach. Learn. Res. Andy Li, Aiden Durrant, Milan Markovic, Tianjin Huang, Souvik Kundu, Tianlong Chen, Lu Yin, Georgios Leontidis
2025	J	`jnl`	SEAL: Steerable Reasoning Calibration of Large Language Models for Free. CoRR Runjin Chen, Zhenyu Zhang, Junyuan Hong, Souvik Kundu, Zhangyang Wang
2025	A*	`conf`	Scaling Long Context Training Data by Long-Distance Referrals. ICLR Yonghao Zhuang, Lanxiang Hu, Longfei Yun, Souvik Kundu, Zhengzhong Liu, Eric P. Xing, Hao Zhang
2025	J	`jnl`	Understanding and Optimizing Multi-Stage AI Inference Pipelines. CoRR Abhimanyu Rajeshkumar Bambhaniya, Hanjiang Wu, Suvinay Subramanian, Sudarshan Srinivasan, Souvik Kundu, Amir Yazdanbakhsh, Midhilesh Elavazhagan, Madhu Kumar, Tushar Krishna
2025	J	`jnl`	Unraveling Zeroth-Order Optimization through the Lens of Low-Dimensional Structured Perturbations. CoRR Sihwan Park, Jihun Yun, Sungyub Kim, Souvik Kundu, Eunho Yang
2024	Misc	`conf`	Analyzing Adversarial Vulnerabilities of Graph Lottery Tickets. ICASSP Subhajit Dutta Chowdhury, Zhiyu Ni, Qingyuan Peng, Souvik Kundu, Pierluigi Nuzzo
2024	J	`jnl`	AttentionBreaker: Adaptive Evolutionary Optimization for Unmasking Vulnerabilities in LLMs through Bit-Flip Attacks. CoRR Sanjay Das, Swastik Bhattacharya, Souvik Kundu, Shamik Kundu, Anand Menon, Arnab Raha, Kanad Basu
2024	J	`jnl`	Bit-by-Bit: Investigating the Vulnerabilities of Binary Neural Networks to Adversarial Bit Flipping. Trans. Mach. Learn. Res. Shamik Kundu, Sanjay Das, Sayar Karmakar, Arnab Raha, Souvik Kundu, Yiorgos Makris, Kanad Basu
2024	J	`jnl`	CLAMP-ViT: Contrastive Data-Free Learning for Adaptive Post-Training Quantization of ViTs. CoRR Akshat Ramachandran, Souvik Kundu, Tushar Krishna
2024	J	`jnl`	Demystifying Platform Requirements for Diverse LLM Inference Use Cases. CoRR Abhimanyu Bambhaniya, Ritik Raj, Geonhwa Jeong, Souvik Kundu, Sudarshan Srinivasan, Midhilesh Elavazhagan, Madhu Kumar, Tushar Krishna
2024	A*	`conf`	Fusing Models with Complementary Expertise. ICLR Hongyi Wang, Felipe Maia Polo, Yuekai Sun, Souvik Kundu, Eric P. Xing, Mikhail Yurochkin
2024	—	`conf`	GEAR: An Efficient Error Reduction Framework for KV Cache Compression in LLM Inference. ENLSP Hao Kang, Qingru Zhang, Souvik Kundu, Geonhwa Jeong, Zaoxing Liu, Tushar Krishna, Tuo Zhao
2024	J	`jnl`	GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM. CoRR Hao Kang, Qingru Zhang, Souvik Kundu, Geonhwa Jeong, Zaoxing Liu, Tushar Krishna, Tuo Zhao
2024	—	`conf`	GenQ: Quantization in Low Data Regimes with Generative Synthetic Data. ECCV (13) Yuhang Li, Youngeun Kim, Donghyun Lee, Souvik Kundu, Priyadarshini Panda
2024	A*	`conf`	Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs "Difficult" Downstream Tasks in LLMs. ICML Lu Yin, Ajay Kumar Jaiswal, Shiwei Liu, Souvik Kundu, Zhangyang Wang
2024	J	`jnl`	LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding. CoRR Doohyuk Jang, Sihwan Park, June Yong Yang, Yeonsung Jung, Jihun Yun, Souvik Kundu, Sungyub Kim, Eunho Yang
2024	J	`jnl`	Metron: Holistic Performance Evaluation Framework for LLM Inference Systems. CoRR Amey Agrawal, Anmol Agarwal, Nitin Kedia, Jayashree Mohan, Souvik Kundu, Nipun Kwatra, Ramachandran Ramjee, Alexey Tumanov
2024	J	`jnl`	MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization. CoRR Akshat Ramachandran, Souvik Kundu, Tushar Krishna
2024	A*	`conf`	ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization. NeurIPS Haoran You, Yipin Guo, Yichao Fu, Wei Zhou, Huihong Shi, Xiaofan Zhang, Souvik Kundu, Amir Yazdanbakhsh, Yingyan (Celine) Lin
2024	J	`jnl`	ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization. CoRR Haoran You, Yipin Guo, Yichao Fu, Wei Zhou, Huihong Shi, Xiaofan Zhang, Souvik Kundu, Amir Yazdanbakhsh, Yingyan Lin
2024	—	`conf`	Towards Real-Time LLM Inference on Heterogeneous Edge Platforms. HIPCW Rakshith Jayanth, Neelesh Gupta, Souvik Kundu, Deepak A. Mathaikutty, Viktor K. Prasanna
2024	J	`jnl`	Unveiling Adversarially Robust Graph Lottery Tickets. Trans. Mach. Learn. Res. Subhajit Dutta Chowdhury, Zhiyu Ni, Qingyuan Peng, Souvik Kundu, Pierluigi Nuzzo
2023	A*	`conf`	Don't just prune by magnitude! Your mask topology is a secret weapon. NeurIPS Duc Hoang, Souvik Kundu, Shiwei Liu, Zhangyang Wang
2023	J	`jnl`	Fusing Models with Complementary Expertise. CoRR Hongyi Wang, Felipe Maia Polo, Yuekai Sun, Souvik Kundu, Eric P. Xing, Mikhail Yurochkin
2023	J	`jnl`	Junk DNA Hypothesis: A Task-Centric Angle of LLM Pre-trained Weights through Sparsity. CoRR Lu Yin, Shiwei Liu, Ajay Jaiswal, Souvik Kundu, Zhangyang Wang
2023	A*	`conf`	NeRFool: Uncovering the Vulnerability of Generalizable Neural Radiance Fields against Adversarial Perturbations. ICML Yonggan Fu, Ye Yuan, Souvik Kundu, Shang Wu, Shunyao Zhang, Yingyan Celine Lin
2023	J	`jnl`	NeRFool: Uncovering the Vulnerability of Generalizable Neural Radiance Fields against Adversarial Perturbations. CoRR Yonggan Fu, Ye Yuan, Souvik Kundu, Shang Wu, Shunyao Zhang, Yingyan Lin
2023	J	`jnl`	Sparse but Strong: Crafting Adversarially Robust Graph Lottery Tickets. CoRR Subhajit Dutta Chowdhury, Zhiyu Ni, Qingyuan Peng, Souvik Kundu, Pierluigi Nuzzo
2023	A*	`conf`	Vision HGNN: An Image is More than a Graph of Nodes. ICCV Yan Han, Peihao Wang, Souvik Kundu, Ying Ding, Zhangyang Wang