[Submitted on 7 May 2026 (v1), last revised 22 Oct 2026 (this version, v2)]
Abstract:In Large Language Model (LLM) fine-tuning, parameter and data selection are common strategies for reducing fine-tuning cost, yet they are typically driven by separate scoring mechanisms. When a parameter mask and data subset jointly determine restricted fine-tuning, this separation incurs redundant overhead and makes coordinated selection difficult. We cast parameter and data selection as two bilevel selection problems under a common validation objective and derive a shared local response-surrogate scoring rule. Under first- and second-order validation-improvement approximations, parameter importance and data utility emerge as column-wise and row-wise aggregations of a single gradient interaction matrix, yielding a closed-form row-column correspondence for co-extracting both signals. Building on this structure, we propose DualSFT (Dual-Selection Fine-Tuning), a one-shot dual-scoring algorithm that produces a parameter mask and data subset from shared gradient statistics. On 3B-9B LLMs, single-axis DualSFT variants strengthen target-task performance and stability-plasticity trade-offs within their comparison groups, while full DualSFT yields a more favorable joint-constrained trade-off than sequential hybrid baselines under matched budgets.
Submission history
From: XinRui Chen [view email]
[v1] Thu, 7 May 2026 12:52:02 UTC (4,363 KB)
[v2] Thu, 22 Oct 2026 01:38:46 UTC (1,858 KB)
