Picture for Souvik Kundu

Souvik Kundu

Callie

OuroMamba: A Data-Free Quantization Framework for Vision Mamba Models

Add code
Mar 13, 2025
Viaarxiv icon

Enhancing Large Language Models for Hardware Verification: A Novel SystemVerilog Assertion Dataset

Add code
Mar 11, 2025
Viaarxiv icon

LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model Compression

Add code
Mar 06, 2025
Viaarxiv icon

LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models

Add code
Feb 10, 2025
Figure 1 for LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models
Figure 2 for LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models
Figure 3 for LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models
Figure 4 for LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models
Viaarxiv icon

CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing

Add code
Feb 04, 2025
Viaarxiv icon

Unraveling Zeroth-Order Optimization through the Lens of Low-Dimensional Structured Perturbations

Add code
Jan 31, 2025
Viaarxiv icon

AttentionBreaker: Adaptive Evolutionary Optimization for Unmasking Vulnerabilities in LLMs through Bit-Flip Attacks

Add code
Nov 21, 2024
Viaarxiv icon

MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization

Add code
Nov 08, 2024
Figure 1 for MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization
Figure 2 for MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization
Figure 3 for MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization
Figure 4 for MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization
Viaarxiv icon

LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding

Add code
Oct 04, 2024
Figure 1 for LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
Figure 2 for LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
Figure 3 for LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
Figure 4 for LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
Viaarxiv icon

Understanding the Performance and Estimating the Cost of LLM Fine-Tuning

Add code
Aug 08, 2024
Figure 1 for Understanding the Performance and Estimating the Cost of LLM Fine-Tuning
Figure 2 for Understanding the Performance and Estimating the Cost of LLM Fine-Tuning
Figure 3 for Understanding the Performance and Estimating the Cost of LLM Fine-Tuning
Figure 4 for Understanding the Performance and Estimating the Cost of LLM Fine-Tuning
Viaarxiv icon