Publications

In this paper, we introduce the Tensor Streaming Processor (TSP) architecture, a functionally-sliced microarchitecture with memory units interleaved with vector and matrix deep learning functional units in order to take advantage of dataflow locality of deep learning operations. The TSP is built based on two key observations: (1) machine learning workloads exhibit abundant data parallelism, which can be readily mapped to tensors in hardware, and (2) a simple and deterministic processor with producer-consumer stream programming model enables precise reasoning and control of hardware components, achieving good performance and power efficiency. The TSP is designed to exploit parallelism inherent in machine-learning workloads including instruction-level, memory concurrency, data and model parallelism, while guaranteeing determinism by eliminating all reactive elements in the hardware (e.g. arbiters, and caches). Early ResNet50 image classification results demonstrate 20.4K processed images per second (IPS) with a batch-size of one— a 4× improvement compared to other modern GPUs and accelerators. Our first ASIC implementation of the TSP architecture yields a computational density of more than 1 TeraOp/s per square mm of silicon for its 25×29 mm 14nm chip operating at a nominal clock frequency of 900 MHz. The TSP demonstrates a novel hardware-software approach to achieve fast, yet predictable, performance on machine-learning workloads within a desired power envelope.

The authors introduce ChatEHR, an internally developed, vendor-agnostic system for integrating large language models (LLMs) directly into clinical workflows within an academic medical center's electronic health record. By enabling real-time access to longitudinal patient data through both automated tasks and an interactive chat interface, the platform successfully reduces manual documentation burden and facilitates decision-making. The authors also establish a robust framework for governance, continuous performance monitoring, and value assessment, demonstrating a replicable model for other health systems.

This paper proposes the development of an AI-Driven Digital Organism (AIDO), a comprehensive, multiscale foundation model system designed to simulate biological systems from the molecular level to entire organisms. By integrating multimodal datasets and hierarchical deep learning architectures, AIDO aims to provide a safe, scalable, and programmable environment for biomedical research and drug discovery. The framework establishes a unified blueprint for modeling complex biological networks, bridging gaps between genotype, phenotype, and environmental influences.

quantum-classical supercomputing: quantum chemistry of protein-ligand complexes

Researchers developed a novel stem cell-based protocol to generate 'cardiobots'—self-organizing aggregates capable of muscle-powered motility. By optimizing mesoderm and cardiogenic induction through precise growth factor signaling and synthetic organizers, the team successfully engineered bio-aggregates with enhanced contractile areas and increased motility compared to existing gastruloid models. This approach provides a new framework for creating autonomous, muscle-propelled biological machines and exploring the evolutionary origins of movement in early multicellular organisms.

One of the biggest hurdles robotics faces is the facet of sophisticated and hard-to-engineer behaviors. Reinforcement learning offers a set of tools, and a framework to address this problem. In parall

Introducing new AI model GET (general expression transformer), an interpretable foundation model designed to uncover regulatory grammars across 213 human fetal and adult cell types

The authors introduce ChatEHR, an internally developed, vendor-agnostic system for integrating large language models (LLMs) directly into clinical workflows within an academic medical center's electronic health record. By enabling real-time access to longitudinal patient data through both automated tasks and an interactive chat interface, the platform successfully reduces manual documentation burden and facilitates decision-making. The authors also establish a robust framework for governance, continuous performance monitoring, and value assessment, demonstrating a replicable model for other health systems.

In this paper, we introduce the Tensor Streaming Processor (TSP) architecture, a functionally-sliced microarchitecture with memory units interleaved with vector and matrix deep learning functional units in order to take advantage of dataflow locality of deep learning operations. The TSP is built based on two key observations: (1) machine learning workloads exhibit abundant data parallelism, which can be readily mapped to tensors in hardware, and (2) a simple and deterministic processor with producer-consumer stream programming model enables precise reasoning and control of hardware components, achieving good performance and power efficiency. The TSP is designed to exploit parallelism inherent in machine-learning workloads including instruction-level, memory concurrency, data and model parallelism, while guaranteeing determinism by eliminating all reactive elements in the hardware (e.g. arbiters, and caches). Early ResNet50 image classification results demonstrate 20.4K processed images per second (IPS) with a batch-size of one— a 4× improvement compared to other modern GPUs and accelerators. Our first ASIC implementation of the TSP architecture yields a computational density of more than 1 TeraOp/s per square mm of silicon for its 25×29 mm 14nm chip operating at a nominal clock frequency of 900 MHz. The TSP demonstrates a novel hardware-software approach to achieve fast, yet predictable, performance on machine-learning workloads within a desired power envelope.

The authors introduce ChatEHR, an internally developed, vendor-agnostic system for integrating large language models (LLMs) directly into clinical workflows within an academic medical center's electronic health record. By enabling real-time access to longitudinal patient data through both automated tasks and an interactive chat interface, the platform successfully reduces manual documentation burden and facilitates decision-making. The authors also establish a robust framework for governance, continuous performance monitoring, and value assessment, demonstrating a replicable model for other health systems.