QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning
This work addresses the problem of memory constraints in fine-tuning LLMs for researchers and practitioners, offering a scalable solution that is incremental in nature.
The paper tackles the computational cost of fine-tuning large language models by proposing QLESS, a method that integrates gradient quantization with the LESS framework to enable memory-efficient data valuation and selection, reducing memory usage by up to 16x while maintaining comparable performance.
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.