QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909474300624896 |
|---|---|
| author | Ananta, Moses Adilazuarda, Muhammad Farid Zuhri, Zayd Muhammad Kawakibi Purwarianti, Ayu Aji, Alham Fikri |
| author_facet | Ananta, Moses Adilazuarda, Muhammad Farid Zuhri, Zayd Muhammad Kawakibi Purwarianti, Ayu Aji, Alham Fikri |
| contents | Fine-tuning large language models (LLMs) is often constrained by the computational costs of processing massive datasets. We propose \textbf{QLESS} (Quantized Low-rank Gradient Similarity Search), which integrates gradient quantization with the LESS framework to enable memory-efficient data valuation and selection. QLESS employs a two-step compression process: first, it obtains low-dimensional gradient representations through LoRA-based random projection; then, it quantizes these gradients to low-bitwidth representations. Experiments on multiple LLM architectures (LLaMA, Mistral, Qwen) and benchmarks (MMLU, BBH, TyDiQA) show that QLESS achieves comparable data selection performance to LESS while reducing memory usage by up to 16x. Even 1-bit gradient quantization preserves data valuation quality. These findings underscore QLESS as a practical, scalable approach to identifying informative examples within strict memory constraints. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01703 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning Ananta, Moses Adilazuarda, Muhammad Farid Zuhri, Zayd Muhammad Kawakibi Purwarianti, Ayu Aji, Alham Fikri Machine Learning Artificial Intelligence Computation and Language Fine-tuning large language models (LLMs) is often constrained by the computational costs of processing massive datasets. We propose \textbf{QLESS} (Quantized Low-rank Gradient Similarity Search), which integrates gradient quantization with the LESS framework to enable memory-efficient data valuation and selection. QLESS employs a two-step compression process: first, it obtains low-dimensional gradient representations through LoRA-based random projection; then, it quantizes these gradients to low-bitwidth representations. Experiments on multiple LLM architectures (LLaMA, Mistral, Qwen) and benchmarks (MMLU, BBH, TyDiQA) show that QLESS achieves comparable data selection performance to LESS while reducing memory usage by up to 16x. Even 1-bit gradient quantization preserves data valuation quality. These findings underscore QLESS as a practical, scalable approach to identifying informative examples within strict memory constraints. |
| title | QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2502.01703 |