Compiling agentic workflows into LLM weights creates subterranean agents with near-frontier quality at two orders of magnitude less cost, validated empirically on travel booking, Zoom support, and insurance claims tasks.
ICLR , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.AI 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Adapts conformal prediction methods to provide distribution-free uncertainty quantification and coverage guarantees for continuous evaluation of AI agent quality scores.
citing papers explorer
-
Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost
Compiling agentic workflows into LLM weights creates subterranean agents with near-frontier quality at two orders of magnitude less cost, validated empirically on travel booking, Zoom support, and insurance claims tasks.
-
Distribution-Free Uncertainty Quantification for Continuous AI Agent Evaluation
Adapts conformal prediction methods to provide distribution-free uncertainty quantification and coverage guarantees for continuous evaluation of AI agent quality scores.