CADFT improves supervised fine-tuning of large language models by dynamically down-weighting training samples whose low model-likelihood indicates high gradient variance, yielding better stability and generalization.
Faster Acceleration for Steepest Descent , booktitle =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Compatibility-Aware Dynamic Fine-Tuning for Large Language Models
CADFT improves supervised fine-tuning of large language models by dynamically down-weighting training samples whose low model-likelihood indicates high gradient variance, yielding better stability and generalization.