Updating and pruning merged LoRA modules on small target-task data improves low-resource summarization over frozen-weight LoRA merging baselines.
This combination ensures that the summary captures the main content more comprehensively, as the highlights alone may sometimes lack sufficient detail
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive LoRA Merge with Parameter Pruning for Low-Resource Generation
Updating and pruning merged LoRA modules on small target-task data improves low-resource summarization over frozen-weight LoRA merging baselines.