English

Slimming Down LLMs Without Losing Their Minds

Computation and Language 2025-06-13 v1 Artificial Intelligence

Abstract

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.

Keywords

Cite

@article{arxiv.2506.10885,
  title  = {Slimming Down LLMs Without Losing Their Minds},
  author = {Qingda and Mai},
  journal= {arXiv preprint arXiv:2506.10885},
  year   = {2025}
}

Comments

10 pages

R2 v1 2026-07-01T03:13:52.399Z