DualSFT derives parameter masks and data subsets as row- and column-wise aggregations of one gradient interaction matrix under first- and second-order validation-improvement approximations.
Diversity as a reward: Fine-tuning LLMs on a mixture of domain-undetermined data
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
One Algorithm, Two Goals: Dual Scoring for Parameter and Data Selection in LLM Fine-Tuning
DualSFT derives parameter masks and data subsets as row- and column-wise aggregations of one gradient interaction matrix under first- and second-order validation-improvement approximations.