Search papers, labs, and topics across Lattice.
TRIAGE introduces a three-level routing framework that enhances the efficiency of Large Language Model (LLM) agents by reusing historical execution trajectories, addressing the inefficiency of the ReAct paradigm. By classifying queries into Direct Reuse, Skill Substitution, and Full ReAct, TRIAGE achieves significant token savings—62.3% in security monitoring queries and 76.3% across diverse domains. The method demonstrates a rapid improvement in efficiency, with a cold-start-to-mature evolution that increases the hit rate for skill reuse from 0% to 57% within just 100 queries.
TRIAGE transforms LLM efficiency by turning historical execution into reusable skills, slashing token consumption to zero for many queries.
Large Language Model (LLM) agents based on the ReAct paradigm have demonstrated remarkable capabilities in tool use and task execution. However, ReAct suffers from a fundamental efficiency problem: every query triggers a complete reasoning loop from scratch, and similar queries repeat identical steps without leveraging historical experience. We propose TRIAGE,a three-level routing framework that reduces token consumption by reusing historical execution trajectories. Its core innovation is TaaS (Trajectory-as-a-Skill), which abstracts historical execution trajectories into reusable skills, realizing 'experience as a service'. TRIAGE classifies queries into three levels: (1) Direct Reuse-identical queries, 0 tokens; (2) Skill Substitution-similar queries, 0 tokens via deterministic parameter substitution; (3) Full ReAct-novel queries, automatically stored for future reuse. In large-scale experiments on 1,007 security monitoring queries, TRIAGE achieves 62.3% token savings, with 56.0% of queries at Level 2 and 5.5% at Level 1, both executing at zero cost. Cross-domain validation on ToolBench (15 domains, 345 queries) achieves 76.3% token reduction, confirming the generalizability of semantic routing. An online learning experiment demonstrates cold-start-to-mature evolution: the L2 hit rate rises from 0% to 57% within the first 100 queries, and the average token cost drops from 198 to 74.7. We also propose an automatic Skill extraction mechanism that distills high-frequency trajectory patterns into deterministic Skills, creating a positive feedback loop of 'the more you use it, the more efficient it becomes'.
