Fine-Tuning on Noisy Instructions: Effects on Generalization and Performance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Alajrami, Ahmed, Tan, Xingwei, Aletras, Nikolaos
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917075158564864
author Alajrami, Ahmed
Tan, Xingwei
Aletras, Nikolaos
author_facet Alajrami, Ahmed
Tan, Xingwei
Aletras, Nikolaos
contents Instruction-tuning plays a vital role in enhancing the task-solving abilities of large language models (LLMs), improving their usability in generating helpful responses on various tasks. However, previous work has demonstrated that they are sensitive to minor variations in instruction phrasing. In this paper, we explore whether introducing perturbations in instruction-tuning data can enhance LLMs' resistance against noisy instructions. We focus on how instruction-tuning with perturbations, such as removing stop words or shuffling words, affects LLMs' performance on the original and perturbed versions of widely-used benchmarks (MMLU, BBH, GSM8K). We further assess learning dynamics and potential shifts in model behavior. Surprisingly, our results suggest that instruction-tuning on perturbed instructions can, in some cases, improve downstream performance. These findings highlight the importance of including perturbed instructions in instruction-tuning, which can make LLMs more resilient to noisy user inputs.
format Preprint
id arxiv_https___arxiv_org_abs_2510_03528
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fine-Tuning on Noisy Instructions: Effects on Generalization and Performance
Alajrami, Ahmed
Tan, Xingwei
Aletras, Nikolaos
Computation and Language
Instruction-tuning plays a vital role in enhancing the task-solving abilities of large language models (LLMs), improving their usability in generating helpful responses on various tasks. However, previous work has demonstrated that they are sensitive to minor variations in instruction phrasing. In this paper, we explore whether introducing perturbations in instruction-tuning data can enhance LLMs' resistance against noisy instructions. We focus on how instruction-tuning with perturbations, such as removing stop words or shuffling words, affects LLMs' performance on the original and perturbed versions of widely-used benchmarks (MMLU, BBH, GSM8K). We further assess learning dynamics and potential shifts in model behavior. Surprisingly, our results suggest that instruction-tuning on perturbed instructions can, in some cases, improve downstream performance. These findings highlight the importance of including perturbed instructions in instruction-tuning, which can make LLMs more resilient to noisy user inputs.
title Fine-Tuning on Noisy Instructions: Effects on Generalization and Performance
topic Computation and Language
url https://arxiv.org/abs/2510.03528