Instruction Fine-Tuning: Does Prompt Loss Matter?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huerta-Enochian, Mathew, Ko, Seung Yong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914970943356928
author Huerta-Enochian, Mathew
Ko, Seung Yong
author_facet Huerta-Enochian, Mathew
Ko, Seung Yong
contents We present a novel study analyzing the effects of various prompt loss token weights (PLW) for supervised instruction fine-tuning (SIFT). While prompt-masking (PLW = 0) is common for SIFT, some fine-tuning APIs support fractional PLWs and suggest that using a small non-zero PLW can help stabilize learning when fine-tuning on short-completion data. However, there has never been a study confirming this claim, and OpenAI, a major cloud-based SIFT provider, recently removed this parameter from their fine-tuning API. We found that performance of models fine-tuned on short-completion data had a statistically-significant negative quadratic relationship with PLW. Using small values (0.01 - 0.5) of PLW produced better results on multiple-choice and short-generation benchmarks (outperforming models fine-tuned on long-completion data) while large values (~ 1.0) of PLW produced better results on long-generation benchmarks. We explained this effect and verified its importance through additional experiments. This research serves as a warning to API providers about the importance of providing a PLW parameter for SIFT.
format Preprint
id arxiv_https___arxiv_org_abs_2401_13586
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Instruction Fine-Tuning: Does Prompt Loss Matter?
Huerta-Enochian, Mathew
Ko, Seung Yong
Machine Learning
Artificial Intelligence
Computation and Language
We present a novel study analyzing the effects of various prompt loss token weights (PLW) for supervised instruction fine-tuning (SIFT). While prompt-masking (PLW = 0) is common for SIFT, some fine-tuning APIs support fractional PLWs and suggest that using a small non-zero PLW can help stabilize learning when fine-tuning on short-completion data. However, there has never been a study confirming this claim, and OpenAI, a major cloud-based SIFT provider, recently removed this parameter from their fine-tuning API. We found that performance of models fine-tuned on short-completion data had a statistically-significant negative quadratic relationship with PLW. Using small values (0.01 - 0.5) of PLW produced better results on multiple-choice and short-generation benchmarks (outperforming models fine-tuned on long-completion data) while large values (~ 1.0) of PLW produced better results on long-generation benchmarks. We explained this effect and verified its importance through additional experiments. This research serves as a warning to API providers about the importance of providing a PLW parameter for SIFT.
title Instruction Fine-Tuning: Does Prompt Loss Matter?
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2401.13586