EASy trains an LLM orchestrator with tree-structured rollouts and multi-component rewards to jointly optimize task accuracy and execution cost, outperforming agentic baselines on seven benchmarks.
Title resolution pending
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
-
EASy: Towards Efficient LLM-Based Agentic System
EASy trains an LLM orchestrator with tree-structured rollouts and multi-component rewards to jointly optimize task accuracy and execution cost, outperforming agentic baselines on seven benchmarks.