LogRouter uses a cost-aware two-level router to select among four execution paths and LLM sizes for log QA, reaching 88.4% router accuracy and 55% lower latency than a fixed 32B baseline on LogHub datasets while keeping quality close.
Reciprocal rank fusion outperforms condorcet and individual rank learning methods,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LogRouter: Adaptive Two-Level LLM Routing for Log Question Answering in Big Data Systems
LogRouter uses a cost-aware two-level router to select among four execution paths and LLM sizes for log QA, reaching 88.4% router accuracy and 55% lower latency than a fixed 32B baseline on LogHub datasets while keeping quality close.