A cross-layer optimization method jointly designs chiplet architecture, training strategy, and optical interconnect topology.
Explore technology by industry and role
Newest additions first. Explore past and current notes by industry and role.
All notes
Newest additions first. Explore past and current notes by industry and role.
The first benchmark that combines agent-designed experiments, a deliberately flawed simulator, and calibrated interval scoring against ground truth.
A communication-oriented chiplet is repurposed as an internal observation platform
A model-agnostic framework that provides time-resolved explanations for dynamic survival predictions with longitudinal data.
A learnable query per degree of freedom enables context-adaptive fusion of visual and EMG data.
A global memory layout stores chiplet-local data contiguously, enabling compatibility between locality-aware placement and page-granularity interleaving.
Proposes a dynamic compute relocation method that moves both computation and data context
LLMs directly program statistical computations in tabular foundation models without gradient-based optimization.
A pushforward prior using sigmoid transformation enables efficient representation of piecewise-constant doping profiles with sharp interfaces in nonlinear PDEs
A novel footprint mask unifies area compactness and aspect ratio penalty into a single spatial cost map.
A software-hardware co-designed prefetcher that reduces HBF latency by tracking page demand with demand-paced staging
Introduces Global Regression Pretraining (GREP) to learn temporal patterns from unlabeled sensor signals.
A unified framework jointly considers thermal and area constraints during design space exploration.
An open-weight 27B reasoning model is trained to natively use real robots through simulation-based reinforcement learning.
Dynamically manages hot expert residency across memory tiers to balance compute and communication load.