Lookahead-R: Budget-Aware Tool Retrieval via Execution-Centric Planning

arXiv:2609.35811 · cs.CL, cs.AI, cs.SE · Submitted 2026-09-20 · Read on arXiv

cs.CL, cs.AI, cs.SE

Submitted: 2026-09-20

Updated: 2026-09-20

Comments: 10 pages, 3 figures, 3 tables. Published in ICMR 2026

DOI: 10.1145/3805622.3810632

License: http://creativecommons.org/licenses/by/4.0/

The gist: Tool retrieval is a critical bottleneck for LLM-based agents operating over large, heterogeneous API ecosystems.

Terminology

Abstract

Tool retrieval is a critical bottleneck for LLM-based agents operating over large, heterogeneous API ecosystems. Existing approaches face an inherent trade-off: semantic retrievers are fast but suffer from the semantic-functional gap, while execution-based validation improves precision at the cost of prohibitive latency. We propose Lookahead-R, a planning-based framework that reformulates tool retrieval as a resource-constrained sequential decision-making problem. At its core, Lookahead-R introduces a lightweight execution-aware surrogate world model that jointly predicts tool execution success, latency cost, and semantic utility---without invoking real APIs. This world model drives a cost-sensitive, uncertainty-guided Monte Carlo Tree Search that navigates the tool space under strict budget constraints. Evaluated on the large-scale ToolBench benchmark, Lookahead-R achieves a superior accuracy-efficiency trade-off across all test scenarios. On the most challenging I3 split, it attains an NDCG@5 of 91.40%, outperforming the state-of-the-art ToolGen (90.16%) by 1.24%. Ablation studies confirm that explicit latency modeling is the key discriminative signal for identifying high-quality tools under resource constraints.

Sources

Related papers