11 papers · 1 filter
SpecReuse: Spectral Graph Reuse for Efficient Vision GNN Inference on FPGAs
Isabella Bernhardt Eiliya, Anvitha Ramachandran, Dhruv Parikh +1
Dynamic Image Graph Construction (DIGC) is the primary performance bottleneck in FPGA acceleration of Vision Graph Neural Networks (ViGs), reconstructing graph connectivity at ever…
FPGA Acceleration of Fully Homomorphic Encryption with Adaptive Key Switching
Zhihan Xu, Jayashree Adivarahan, Rajgopal Kannan +1
Fully Homomorphic Encryption (FHE) enables privacy-preserving cloud services but incurs substantial computation overhead, making hardware acceleration essential. Among FHE operatio…
FOLIO: Focused Semantic Memory for Streaming Video Understanding
Haoyang Fan, Dhruv Parikh, Anvitha Ramachandran +4
In online streaming video understanding, a video stream continues to arrive and queries may be issued at any time. Because streaming frames grow without bound, the system must cont…
Vision Non-Causal Trapezoidal Mamba: Eliminating Directional Scanning in Vision SSMs with Second-Order Dynamics
Anvitha Ramachandran, Dhruv Parikh, Haoyang Fan +2
State Space Models (SSMs) have emerged as an alternative to Vision Transformers, yet most vision SSMs inherit directional token scanning from causal sequence modeling. While effect…
GraphLeap: Decoupling Graph Construction and Convolution for Vision GNN Acceleration on FPGA
Anvitha Ramachandran, Dhruv Parikh, Viktor Prasanna
Vision Graph Neural Networks (ViGs) represent an image as a graph of patch tokens, enabling adaptive, feature-driven neighborhoods. Unlike CNNs with fixed grid biases or Vision Tra…
ImageHD: Energy-Efficient On-Device Continual Learning of Visual Representations via Hyperdimensional Computing
Jebacyril Arockiaraj, Dhruv Parikh, Viktor Prasanna
On-device continual learning (CL) is critical for edge AI systems operating on non-stationary data streams, but most existing methods rely on backpropagation or exemplar-heavy clas…
Latent Denoising Improves Visual Alignment in Large Multimodal Models
Dhruv Parikh, Jacob Fein-Ashley, Rajgopal Kannan +1
Large Multimodal Models (LMMs) such as LLaVA are typically trained with an autoregressive language modeling objective, providing only indirect supervision to visual tokens. This of…
A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGA
Neelesh Gupta, Peter Wang, Rajgopal Kannan +1
Gated DeltaNet (GDN) is a linear attention mechanism that replaces the growing KV cache with a fixed-size recurrent state. Hybrid LLMs like Qwen3-Next use 75% GDN layers and achiev…
ARMOR: Robust and Efficient CNN-Based SAR ATR through Model-Hardware Co-Design
Sachini Wickramasinghe, Tian Ye, Cauligi Raghavendra +1
Convolutional Neural Networks (CNNs) have achieved state-of-the-art accuracy in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR). However, their high computational…
FAST-Prefill: FPGA Accelerated Sparse Attention for Long Context LLM Prefill
Rakshith Jayanth, Viktor Prasanna
In long-context large language model (LLM) inference, the prefill stage dominates computation due to self-attention over the complete input context. Sparse attention significantly…
Primitive-Driven Acceleration of Hyperdimensional Computing for Real-Time Image Classification
Dhruv Parikh, Jebacyril Arockiaraj, Viktor Prasanna
Hyperdimensional Computing (HDC) represents data using extremely high-dimensional, low-precision vectors, termed hypervectors (HVs), and performs learning and inference through lig…