ISCAS 2025

200 papers
The 2025 edition of the International Symposium on Circuits and Systems (ISCAS) featured 191 papers focusing on the intersection of hardware architecture and advanced computational methods. Key research directions included bridging volumetric density with explicit surface geometry and applying physics-informed deep learning to underwater machine vision. Prolific authors contributing to the proceedings included Fan Zhang, Guangtao Zhai, and Xiongkuo Min. Research themes emphasized hardware-algorithm co-design, specifically through mixed-alphabet set multipliers for Deep Neural Network (DNN) workloads and the analysis of capacitor mismatch nonlinearity in analog compute-in-memory (CIM) architectures. Technical developments also addressed combinatorial optimization using ferroelectric capacitor-based nvSRAM and medical applications such as attention-based CNNs for image classification and gamified interactive learning for cardiopulmonary sound labeling. Other work explored robust parallel operations for electric drive shaft systems and linear-regression-assisted trimming schemes for CMOS voltage references. These contributions reflect a focus on integrating specialized hardware with machine learning and signal processing applications.

Top Research Directions

Research directions most represented by ISCAS 2025 papers.

Learning to Design Hardware Under Extreme Data Scarcity 2
Perceptual Quality Assessment for Stereoscopic Media 1
Implicit Neural Representations for Video Compression 1
Data Summarization via Sensitivity Sampling 1
Hybrid Pointwise Pairwise Optimization for Robust Ranking 1
Volumetric Surface Reconstruction via Depth Map Fusion 1
Resolving Superposition with Sparse Autoencoders 1
Overcoming Data Conversion Bottlenecks in Analog Computing 1
Resolving Surface Ambiguity in Volumetric Neural Fields 1
Verifying Synthetic Data Through Code Execution 1
Fractional Bit Width Quantization for Neural Networks 1
Volumetric Label Denoising via Multi-View Spatial Consensus 1
Mitigating Spectral Bias with Fourier Feature Mapping 1
Robust Deep Learning on Noisy Analog Memory Crossbars 1
Mitigating the Computational Overhead of 4D Light Fields 1

Top Authors

AuthorPapers
Fan Zhang5
Jie Yang5
Xiongkuo Min4
David Bull3
Davide Zoni3
Guangtao Zhai3
Jieming Yin3
Samuel Yen-Chi Chen3
Sergi Abadal3
Zhibo Chen3
Ahmet Enis Cetin2
Arindam Basu2
Chang Liu2
De Ma2
Fanyang Meng2
Huazhong Yang2
Johann Knechtel2
Julian Büchel2
Li Li2
Manuel Le Gallo2

Impactful Papers

Papers from ISCAS 2025 that shaped research clusters.

Influential

Papers

Title
Educational Framework for Power Side-Channel Attacks on Neural Networks in Embedded Systems
EEPNet: Efficient Edge Pixel-based Matching Network for Cross-Modal Dynamic Registration between LiDAR and Camera
Effect of Capacitor Mismatch Nonlinearity on Inference Accuracy in Analog Compute-in-Memory Architectures
Efficient Bearing Sensor Data Compression via an Asymmetrical Autoencoder with a Lifting Wavelet Transform Layer
Efficient End-to-End 6-Dof Grasp Detection Framework for Edge Devices with Hierarchical Heatmaps and Feature Propagation
Efficient Image Compression through Extreme Image Rescaling
Efficient Lossless Compression with Distribution Quantized Finite-State Autoregressive Model
Efficient Real-Time Fine-Grained Action Recognition over a Progressive and Hierarchical Classification Framework
Efficient Self-Adaptive Pseudo-Resistor with Rapid Settling and High Linearity for Neurorecording Front-End Circuits
EFVC: Error-Propagation-Free Neural Video Coding with Reversible Transform
Enabling Fault Isolation in Fault Injection for Automotive-grade Circuits Simulation
Enhancing HDR Video Compression based on Deep Effective Bit Depth Adaptation
Enhancing LLM Performance on Hardware Design Generation Task via Reinforcement Learning
Event-based Neural Spike Detection Using Spiking Neural Networks for Neuromorphic iBMI Systems
FARMER: An Online-Learning Driven Methodology for Workload Consolidation on Large FPGAs
Fast and High-Precision Analog In-Sensor Visual Computing Using Fully Amorphous Metal Oxide Thin-Film Transistors
FastViT: Real-Time Linear Attention Accelerator for Dense Predictions of Vision Transformer (ViT)
Feature Perturbation Agent based Adversarial Attack Method for Weakly Supervised Video Anomaly Detection
Fixed-Throughput GRAND with FIFO Scheduling
FlexAcc: Accelerating Batch Normalization through GPU-FPGA Integration
Fourier Phase Retrieval With Diffusion Priors Guided by Failure Detection
FPGA Accelerated Adaptive LDPC-based Quantum Error Correction by Bitwise Pipeline Parallelism
FrameVoting: A Robust and Fast Method of Using Gaze Estimations to Identify Objects of Interest
Function-Reused Oscillator-PA for IoT Applications
Gain Cell-based Analog Content Addressable Memory for Dynamic Associative Tasks in AI
Genetic Motifs as a Blueprint for Mismatch-Tolerant Neuromorphic Computing
GraphDiffusion: A Graph-conditioned Diffusion Model for Chip Placement
Grouped Transform for Ultra-Low-Complexity Learned Image Compression
Harnessing Forecast Uncertainty in Deep Learning for Time Series Anomaly Detection with Posterior Distribution Scoring
HEALSound: Healthcare Education And Labeling for Cardiopulmonary Sounds
HieRFP: A Hierarchical Recognition and Floorplanning Framework for Reusable Modules
High-Fidelity Object Removal through Boosting Diffusion Processes
High Time-Resolution PWM generator for GaN Converters
HRHuman: Tuning-Free Higher-Resolution Human Image Generation via Template Knowledge
HSCIM: A High Security Compute-In-Memory Architecture with PUF based on TST-MRAM
Input-adaptive Mixed-Precision Framework for Efficient Object Detection
InternVQA: Advancing Compressed Video Quality Assessment with Distilling Large Foundation Model
Introduction to Quantum Machine Learning and Quantum Architecture Search
Ising machine based on charge re-distribution
ITERTL: An Iterative Framework for Fine-tuning LLMs for RTL Code Generation
IVCA: Inter-relation-aware Video Complexity Analyzer
Jumping Memory for Memristive Reservoir Computing
Kung-Fu: An Energy-Efficient Compute-In-Memory Approach for Neural Network Inference Using Multi-Level Binary Computing Fusion
LayoutCopilot: LLM-Empowered Analog Layout Design towards Enhanced Human-Machine Interaction
LiCSPA: Lightweight Countermeasure against Static Power Side-Channel Attacks
LineShield - A Generalized LiDAR Pipeline for Automated Vegetation Encroachment Detection on Powerlines
Live Demonstration: A real-time event encoder for seizure monitoring on Neuromorphic Hardware
Live Demonstration: Automated DNN Deployment on the IBM HERMES Project Chip
Live Demonstration: Crowdsourcing Cardiopulmonary Sound Labeling via Gamified Interactive Learning (HEALSound)
Live Demonstration: Improving efficiency of speech recognition with neuro-inspired units on AIU Spyre

« prev Page 3/4 next »