Slimming Down LLMs Without Losing Their Minds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qingda, Mai
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911002252017664
author Qingda
Mai
author_facet Qingda
Mai
contents This paper investigates and validates the impact of fine-tuning on large language model performance, focusing on parameter-efficient methods (LoRA and QLoRA). We evaluate model capabilities across three key domains: (1) commonsense reasoning (HellaSwag), (2) mathematical reasoning (GSM8K), and (3) multi-domain knowledge (MMLU-CS). Our findings demonstrate that: (1) LoRA-based methods effectively improve task-specific performance while maintaining computational efficiency, and (2) performance strongly depends on alignment between fine-tuning dataset and benchmark tasks. The study provides both theoretical insights into parameter-efficient mechanisms and practical guidance for developers implementing efficient LLM adaptation with limited resources.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10885
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Slimming Down LLMs Without Losing Their Minds
Qingda
Mai
Computation and Language
Artificial Intelligence
This paper investigates and validates the impact of fine-tuning on large language model performance, focusing on parameter-efficient methods (LoRA and QLoRA). We evaluate model capabilities across three key domains: (1) commonsense reasoning (HellaSwag), (2) mathematical reasoning (GSM8K), and (3) multi-domain knowledge (MMLU-CS). Our findings demonstrate that: (1) LoRA-based methods effectively improve task-specific performance while maintaining computational efficiency, and (2) performance strongly depends on alignment between fine-tuning dataset and benchmark tasks. The study provides both theoretical insights into parameter-efficient mechanisms and practical guidance for developers implementing efficient LLM adaptation with limited resources.
title Slimming Down LLMs Without Losing Their Minds
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2506.10885