Oct 25, 2023 · In this work, we utilize Large Language Models (LLMs) for a novel use case: constructing Performance Predictors (PP) that estimate the performance of specific ...

In this work, we utilize Large Language Models (LLMs) for a novel use case: constructing Performance Predictors (PP) that estimate the performance of specific ...

It reduces search hours by approximately 50% and offers potential improvements in latency, GFLOPs, and model size. This can be a substantial advantage for ...

Aug 11, 2024 · This study explores a unique application of LLMs: constructing a performance predictor (LLM-PP) for a deep neural network (DNN) architecture.

Aug 7, 2024 · The aim is to create a performance predictor with low prediction errors compared to training from scratch. The hypothesis is that LLMs possess a ...

The main goal of this work is to unearth the architecture understanding capability of LLMs to design PPs that are: (i) accurate, (ii) efficient, and (iii) ...

Sep 23, 2024 · Neural architecture search (NAS) and network pruning are widely studied efficient AI techniques, but not yet perfect.

In this work, we utilize Large Language Models (LLMs) for a novel use case: constructing Performance Predictors (PP) that estimate the performance of specific ...

LLM Performance Predictors are good initializers for Architecture Search. This repository contains the code and the data used in the LLM-PP work. This ...

In this work, we utilize Large Language Models (LLMs) for a novel use case: constructing Performance Predictors (PP) that estimate the performance of specific ...