Models trained with primitive awareness achieve task transfer with only three demonstrations
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.
A transparent cache pool and cache streamer efficiently manage GPU memory allocation and cache residency across layers.
An automated framework that revises games based on agent play experience to enhance mastery
A multi-objective reinforcement learning policy that jointly optimizes performance, energy, and thermal constraints in heterogeneous PIM systems.
A novel method efficiently computes feasible stochastic optimal solutions with bounded constraint violation
High-temperature distribution alignment leverages imperfect predictions to expand training data
A mode-world weighted regression loss and iterative decoder are proposed to simultaneously improve mode diversity and prediction accuracy.
MiLSD proposes a new architecture enabling accurate line segment detection within ultra-low memory constraints.
A team of net-carrying drones uses competitive multi-agent reinforcement learning to intercept agile targets
A new method using quantum Hall plateau width to assess disorder in semiconductor heterostructures is proposed.
AI agents can explore multiple design options, integrate them, and iteratively improve designs without fixed design blocks.
A new methodology direction is emerging where AI agents integrate into early design stages of chip development.
A hybrid chiplet architecture integrating analog, digital CIM, and intermediate data processing units for flexible resource scaling.
By early stopping and appending high-confidence base-only categories, object coverage loss in preference-optimized models is restored without training
A computation execution graph-based mapping scheme decouples micro-batch and layer dimensions, enabling fine-grained control on heterogeneous chiplets