Merge-Bench dataset and LLMergeJ model demonstrate that a 14B-parameter LLM trained with GRPO outperforms some commercial models on Java merge resolution but all tested models resolve under 60% of conflicts across 11 languages.
https://arxiv.org/abs/2505.22583 (May 2025)
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
-
Merge-Bench: Resolve Merge Conflicts with Large Language Models
Merge-Bench dataset and LLMergeJ model demonstrate that a 14B-parameter LLM trained with GRPO outperforms some commercial models on Java merge resolution but all tested models resolve under 60% of conflicts across 11 languages.