Razvan-Gabriel Dumitru
- Layer-Wise Quantization: A Pragmatic and Effective Method for Quantizing LLMs Beyond Integer Bit-Levels
2024/06/25 by Razvan-Gabriel Dumitru, Vikas Yadav, Dumitru, Razvan-Gabriel +9 · 5 citations
Engineering · #Advancements in Photolithography Techniques #Advancements in Semiconductor Devices and Circuit Design #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #I.2.0 #I.2.7 #Machine Learning (cs.LG) #Semiconductor materials and devices
- CopySpec: Accelerating LLMs with Speculative Copy-and-Paste Without Compromising Quality
2025/02/13 by Razvan-Gabriel Dumitru, Minglai Yang, Dumitru, Razvan-Gabriel +5 · 2 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Digital Rights Management and Security #FOS: Computer and information sciences #I.2.0 #I.2.7 #Machine Learning (cs.LG) #Semantic Web and Ontologies
- Change Is the Only Constant: Dynamic LLM Slicing based on Layer Redundancy
2024/11/05 by Razvan-Gabriel Dumitru, Dumitru, Razvan-Gabriel, Paul-Ioan Clotan +7 · 1 citation
Computer Science · #Natural Language Processing Techniques #Network Packet Processing and Optimization
- CRAFT: Clustering Rubrics to Diagnose Weak LLM Capabilities and Generate Targeted Fine-Tuning Data
2026/07/17 by Vipul Gupta, Zihao Wang, Razvan-Gabriel Dumitru +3 · 1 voice
#cs.AI #cs.LG