Chip Industry Technical Paper Roundup: August 5

Photonic-integrated circuit packaging; V2X communication; efficient LLM inference; statistical BER analysis for chiplets; LLM 3DHI; preventing end-to-end slowdowns in accelerated chip multi-processors; thermal nanoimprint-based dielectric patterning; scalable neuromorphic computing; intrusion detection of DoS attacks on CAN bus.

popularity

New technical papers recently added to Semiconductor Engineering’s library:

Name of Paper Research Organizations
Advanced Optical Integration Processes for Photonic-Integrated Circuit Packaging Hanyang University
Efficient LLM Inference: Core Bottlenecks Imposed By Memory, Compute Capacity, Synchronization Overheads are all you need NVIDIA
Fast and Accurate Jitter Modeling for Statistical BER Analysis for Chiplet Interconnect and Beyond TSMC
A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration Georgia Tech
RACER: Avoiding End-to-End Slowdowns in Accelerated Chip Multi-Processors Cornell University, Intel Labs
Hybrid Bonding With Polymeric Interlayer Dielectric Layers Patterned by Nanoimprint Lithography Seoul National University of Science and Technology
Neuromorphic Computing: A Theoretical Framework for Time, Space, and Energy Scaling Sandia National Laboratories
CANDoSA: A Hardware Performance Counter-Based Intrusion Detection System for DoS Attacks on Automotive CAN bus Dumarey Softronix, Politecnico di Torino
Research Challenges and Progress in the End-to-End V2X Cooperative Autonomous Driving Competition Tsinghua, HK University, Stanford, TU Munich et al.


Find more semiconductor research papers here.



Leave a Reply


(Note: This name will be displayed publicly)