Parker Hill

TL;DR: A new dataset is introduced that includes queries that are out-of-scope—i.e., queries that do not fall into any of the system’s supported intents, posing a new challenge because models cannot assume that every query at inference time belongs to a system-supported intent class.

TL;DR: DeftNN is a GPU DNN execution framework that targets the key architectural bottlenecks of DNNs on GPUs to automatically and transparently improve execution performance, and is composed of two novel optimization techniques– synapse vector elimination, a technique that identifies non-contributing synapses in the DNN and carefully transforms data and removes the computation and data movement.

TL;DR: This paper introduces Input Responsive Approximation, an approach that uses a canary input — a small program input carefully constructed to capture the intrinsic properties of the original input — to automatically control how program approximation is applied on an input-by-input basis.

TL;DR: The authors introduce a new dataset that includes queries that are out-of-scope, i.e., queries that do not fall into any of the system's supported intents, and evaluate a range of benchmark classifiers on this dataset along with several different out-ofthe-scope identification schemes.

TL;DR: This paper introduces a novel approximate computing technique that decouples the format of data in the memory hierarchy from the format in the compute subsystem to significantly reduce the cost of storing and moving bits throughout theMemory hierarchy and improve application performance.