Surprise-Guided MergeSort uses a VLM-based composite surprise scorer to prioritize human comparisons inside a MergeSort scheduler, skipping up to 535 pairs per session and raising Kendall's τ by 6-12 points over Active Elo at fixed budget across six benchmarks.
arXiv preprint arXiv:2505.24643 (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
-
Surprise-Guided MergeSort: Budget-Efficient Human-in-the-Loop Ranking via Adaptive Comparison Scheduling
Surprise-Guided MergeSort uses a VLM-based composite surprise scorer to prioritize human comparisons inside a MergeSort scheduler, skipping up to 535 pairs per session and raising Kendall's τ by 6-12 points over Active Elo at fixed budget across six benchmarks.