PEFT-Bench: A Parameter-Efficient Fine-Tuning Methods Benchmark

TL;DR

PEFT-Bench provides a comprehensive benchmark for evaluating parameter-efficient fine-tuning methods on large language models across multiple datasets, considering various efficiency metrics.

Contribution

It introduces PEFT-Bench, a unified benchmark with the PSCP metric to evaluate diverse PEFT methods on autoregressive LLMs comprehensively.

Findings

Demonstrates PEFT-Bench's application on 27 NLP datasets.

Introduces the PSCP metric for holistic efficiency evaluation.

Evaluates 7 different PEFT methods.

Abstract

Despite the state-of-the-art performance of Large Language Models (LLMs) achieved on many tasks, their massive scale often leads to high computational and environmental costs, limiting their accessibility. Parameter-Efficient Fine-Tuning (PEFT) methods address this challenge by reducing the number of trainable parameters while maintaining strong downstream performance. Despite the advances in PEFT methods, current evaluations remain limited (in terms of evaluated models and datasets) and difficult to reproduce. To bridge this gap, we introduce PEFT-Bench, a unified end-to-end benchmark for evaluating diverse PEFT methods on autoregressive LLMs. We demonstrate its usage across 27 NLP datasets and 7 PEFT methods. To account for different PEFT training and inference factors, we also introduce the PEFT Soft Cost Penalties (PSCP) metric, which takes trainable parameters, inference speed, and training memory usage into account.