Balancing performance and efficiency in long-horizon agentic search

Determine effective strategies to balance long-horizon agentic search performance and computational efficiency for tool-augmented large language model research agents that conduct multi-step web search, browsing, and evidence aggregation under constrained interaction budgets and latency requirements.

Background

The paper observes that many deep research agents improve results by increasing reasoning depth and the number of tool calls, which in turn drives up inference latency and computational cost in search-intensive scenarios. This tension is especially acute for long-horizon tasks that require extensive external evidence acquisition and verification.

While the authors propose the SMTL framework to address efficiency via parallel evidence acquisition and structured context management, they explicitly note that the broader challenge of balancing task performance with compute and latency constraints for long-horizon agentic search remains unsettled.

References

Balancing long-horizon search performance and computational efficiency remains an open problem.

— Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization  (2602.22675 - Chen et al., 26 Feb 2026) in Section 1 (Introduction)

Some routing anomaly decisions may require information beyond the evidence currently indexed by RouteLLM. An agent-based extension could query additional sources, such as WHOIS, looking glasses, and operator reports. Such an approach may provide richer evidence for complex incidents, but also introduces challenges in tool selection, external-service availability, and inference latency. We leave this exploration for future work.

— The Surprising Effectiveness of LLMs in BGP Security: Mining An Unprecedented Amount of Incidents and Boosting Anomaly Detection  (2608.22812 - Liu et al., 24 Aug 2026) in Section 6, Limitations and Future Work, subsection “Agent-based evidence collection”