1 citations · 3 across the 19 of their papers we have counts for
22 papers
Structure-augmented LLMs for High-Level Synthesis Pragma Optimization
Haocheng Xu, Ye Qiao, Phyo Pyae Moe Aung +7
Pragma insertion drives the quality of high-level synthesis (HLS) designs. Choosing the right directives demands expert knowledge and reasoning about loop nesting, data dependences…
EdgeDAE: Acceleration of Diffusion Action Experts for Real-Time Physical AI with Tiny VLAs on Edge FPGA-GPU Systems
Zhiheng Chen, Ye Qiao, Mohammad Abdullah Al Faruque +1
Physical AI models such as Vision-Language-Action (VLA) architectures enable generalist robotic policies through large-scale transformer backbones and diffusion-based action decode…
Bern2Edge: A Neurosymbolic Compiler for Edge Deployment via Bernstein Polynomial Networks
Malak Gamal El-Din, Yifan Zhang, Yasser Shoukry +2
Deploying high-accuracy neural networks on resource-constrained edge devices remains challenging, as existing approaches treat training, compression, and hardware synthesis as sepa…
LowRank-SSM: Hardware-Software Co-Design for Rank-Reduced Mamba Acceleration on FPGA
Haocheng Xu, Bhardwaj Bhat, Yu-an Chou +6
State Space Models (SSMs) such as Mamba and Mamba-2 achieve linear-time autoregressive inference, making them attractive for latency-sensitive and resource-constrained deployment.…
MOSAIC: Efficient Mixture-of-Agent Scheduling via Adaptive Aggregation and Inference Concurrency
Saptarshi Mitra, Yifan Zhang, Rachid Karami +5
Mixture-of-Agents (MoA) systems improve reasoning accuracy by routing each query to multiple expert LLMs and aggregating their outputs. Efficiently executing this workload on limit…
FASQ: Flexible Accelerated Subspace Quantization for Calibration-Free LLM Compression
Ye Qiao, Yian Wang, Zhiheng Chen +2
Compressing large language models (LLMs) for deployment on commodity GPUs remains challenging: conventional scalar quantization is limited to fixed bit-widths (e.g., 8/4/3-bit), of…
PD-Swap: Prefill-Decode Logic Swapping for End-to-End LLM Inference on Edge FPGAs via Dynamic Partial Reconfiguration
Yifan Zhang, Zhiheng Chen, Ye Qiao +1
Aggressively quantized large language models (LLMs), such as BitNet-style 1.58-bit Transformers with ternary weights, make it feasible to deploy generative AI on low-power edge FPG…
TeLLMe v2: An Efficient End-to-End Ternary LLM Prefill and Decode Accelerator with Table-Lookup Matmul on Edge FPGAs
Ye Qiao, Zhiheng Chen, Yifan Zhang +2
With the emergence of wearable devices and other embedded systems, deploying large language models (LLMs) on edge platforms has become an urgent need. However, this is challenging…
Rethinking RoPE Scaling in Quantized LLM: Theory, Outlier, and Channel-Band Analysis with Weight Rescaling
Ye Qiao, Haocheng Xu, Xiaofan Zhang +1
Extending the context window support of large language models (LLMs) is crucial for tasks with long-distance dependencies. RoPE-based interpolation and extrapolation methods, such…
Q-ROAR: Outlier-Aware Rescaling for RoPE Position Interpolation in Quantized Long-Context LLMs
Ye Qiao, Sitao Huang
Extending LLM context windows is crucial for long range tasks. RoPE-based position interpolation (PI) methods like linear and frequency-aware scaling extend input lengths without r…
DPad: Efficient Diffusion Language Models with Suffix Dropout
Xinhua Chen, Sitao Huang, Cong Guo +5
Diffusion-based Large Language Models (dLLMs) parallelize text generation by framing decoding as a denoising process, but suffer from high computational overhead since they predict…
A Sparsity-Aware Autonomous Path Planning Accelerator with HW/SW Co-Design and Multi-Level Dataflow Optimization
Yifan Zhang, Xiaoyu Niu, Hongzheng Tian +4
Path planning is critical for autonomous driving, generating smooth, collision-free, feasible paths based on perception and localization inputs. However, its computationally intens…
Fault-Free Analog Computing with Imperfect Hardware
Zhicheng Xu, Jiawei Liu, Sitao Huang +9
The growing demand for edge computing and AI drives research into analog in-memory computing using memristors, which overcome data movement bottlenecks by computing directly within…
Characterizing State Space Model and Hybrid Language Model Performance with Long Context
Saptarshi Mitra, Rachid Karami, Haocheng Xu +2
Emerging applications such as AR are driving demands for machine intelligence capable of processing continuous and/or long-context inputs on local devices. However, currently domin…
COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference
Ye Qiao, Zhiheng Chen, Yian Wang +3
Transformer-based models have demonstrated superior performance in various fields, including natural language processing and computer vision. However, their enormous model size and…
TeLLMe: An Energy-Efficient Ternary LLM Accelerator for Prefilling and Decoding on Edge FPGAs
Ye Qiao, Zhiheng Chen, Yifan Zhang +2
Deploying large language models (LLMs) on edge platforms is challenged by their high computational and memory demands. Although recent low-bit quantization methods (e.g., BitNet, D…
Optimized Spatial Architecture Mapping Flow for Transformer Accelerators
Haocheng Xu, Faraz Tahmasebi, Ye Qiao +3
Recent innovations in Transformer-based large language models have significantly advanced the field of general-purpose neural language understanding and generation. With billions o…
Optimizing High-Level Synthesis Designs with Retrieval-Augmented Large Language Models
Haocheng Xu, Haotian Hu, Sitao Huang
High-level synthesis (HLS) allows hardware designers to create hardware designs with high-level programming languages like C/C++/OpenCL, which greatly improves hardware design prod…
MONAS: Efficient Zero-Shot Neural Architecture Search for MCUs
Ye Qiao, Haocheng Xu, Yifan Zhang +1
Neural Architecture Search (NAS) has proven effective in discovering new Convolutional Neural Network (CNN) architectures, particularly for scenarios with well-defined accuracy opt…
RSEND: Retinex-based Squeeze and Excitation Network with Dark Region Detection for Efficient Low Light Image Enhancement
Jingcheng Li, Ye Qiao, Haocheng Xu +1
Images captured under low-light scenarios often suffer from low quality. Previous CNN-based deep learning methods often involve using Retinex theory. Nevertheless, most of them can…
TG-NAS: Generalizable Zero-Cost Proxies with Operator Description Embedding and Graph Learning for Efficient Neural Architecture Search
Ye Qiao, Jingcheng Li, Haocheng Xu +1
Neural Architecture Search (NAS) is a powerful technique for discovering high-performing CNN architectures, but most existing methods rely on costly training or extensive sampling.…
MicroNAS: Zero-Shot Neural Architecture Search for MCUs
Ye Qiao, Haocheng Xu, Yifan Zhang +1
Neural Architecture Search (NAS) effectively discovers new Convolutional Neural Network (CNN) architectures, particularly for accuracy optimization. However, prior approaches often…