News

Congratulations! We have a paper accepted to Symposium on European Solid-State Electronics Research Conference(ESSERC), 2025 “D3TA: 38.9TOPS/W Transformer Accelerator with Dual-Port 3T-eDRAM Digital Compute-In-Memory Using HyperAttention and Triple-Sparsity-Handling”

Congratulations! We have a paper accepted to IEEE Transactions on Computers (TC), 2025 “SAL-PIM: A Subarray-level Processing-in-Memory Architecture with LUT-based Linear Interpolation for Transformer-based Text Generation”

Congratulations! We have a paper accepted to IEEE Symposium on VLSI Technology and Circuits (VLSI), 2025 “A 4nm LLM Accelerator with Streamlined Dataflow and Dual-Mode Parallelization for Efficient Generative AI Inference”

Congratulations! We have a paper accepted to ACM/IEEE International Symposium on Computer Architecture (ISCA), 2025 “Hybe: GPU-NPU Hybrid System for Efficient LLM Inference with Million-Token Context Window”

Congratulations! We have a paper accepted to ACM/IEEE International Symposium on Computer Architecture (ISCA), 2025 “Oaken: Fast and Efficient LLM Serving with Online-Offline Hybrid KV Cache Quantization”

Congratulations! We have a paper accepted to ACM/IEEE International Symposium on Computer Architecture (ISCA), 2025 “Improving Sequence Length Limitation in Protein Structure Prediction Model via Adaptive Activation Quantization”

Congratulations! We have a paper accepted to IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2025 “ADOR: A Design Exploration Framework for LLM Serving with Enhanced Latency and Throughput”

Congratulations! We have a paper accepted to ACM/IEEE Design Automation Conference (DAC), 2025 “ABC-FHE: A Resource-Efficient Accelerator Enabling Bootstrappable Parameters for Client-Side Fully Homomorphic Encryption”

Congratulations! We have a paper accepted to AAAI Conference on Artificial Intelligence(AAAI), 2025 “AoP-SAM: Automation of Prompts for Efficient Segmentation”

Congratulations! We have a paper accepted to IEEE Journal of Solid-State Circuits (JSSC), 2025 “Hawkeye: A Point Cloud Neural Network Processor with Virtual Pillar and Quadtree-based Workload Management for Real-Time Outdoor BEV Detection”