Research map / Large language models / Alignment and preferences
Meta-Router: Bridging Gold-standard and Preference-based Evaluations in LLM Routing
Yichi Zhang, Fangzheng Xie, Shu Yang, Chong Wu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use R- or DR-learner meta-router to debias preference data instead of pooling GS and PB data directly | router trained on GS plus PB data · HealthBench | efficiency gain over random routing | Improved |
| replace random forest with XGBoost as the regressor (ablation) | meta-router · HealthBench | efficiency gain | Mixed |
| add linear mean-shift debiasing of PB data instead of CATE-based debiasing (ablation) | pooled router · HealthBench | efficiency gain | No clear effect |
| normalize GS outcomes to match PB variance (ablation) | meta-router · HealthBench | efficiency gain | Improved |
| PCA dimension 100 instead of 50 (ablation) | meta-router · HealthBench | efficiency gain | No clear effect |
| use Grok 4 Fast as the PB judge instead of GPT-5-mini (ablation) | meta-router · HealthBench | efficiency gain | No clear effect |
| increase GS sample size n (ablation) | routers using GS data · HealthBench | efficiency gain: n = 100, 500, 1000 | Improved |
| use DR-learner rather than R-learner under small samples (ablation) | meta-router · PRBench (676 single-turn questions) | efficiency gain | Improved |
Closest papers by meaning
- RouteLLM: Learning to Route LLMs from Preference Data (ICLR 2025 · similarity 0.93)
- Causal LLM Routing: End-to-End Regret Minimization from Observational Data (NeurIPS 2025 · similarity 0.90)
- Confidence-Guided Stepwise Model Routing for Cost-Efficient Reasoning (AAAI 2026 · similarity 0.90)
- GraphRouter: A Graph-based Router for LLM Selections (ICLR 2025 · similarity 0.89)
- Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Smoothie: Label Free Language Model Routing (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
