STEAM is a training-free test-time framework that improves success rate, makespan, and cost of existing learning-based decentralized MAPF policies by up to 60% via congestion-aware cost-to-go and logit adjustments.
Lacam: Search-based algorithm for quick multi- agent pathfinding
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
STEAM: A Training-Free Congestion-Aware Enhancement Framework for Decentralized Multi-Agent Path Finding
STEAM is a training-free test-time framework that improves success rate, makespan, and cost of existing learning-based decentralized MAPF policies by up to 60% via congestion-aware cost-to-go and logit adjustments.