Pandora’s Router weighs cost and quality in model selection
Published on August 22, 2026, the paper frames routing as a decision about when better estimates are worth paying for. On MATH, RAG, and EmbedLLM, it achieved or tied for the lowest average combined inspection cost and routing regret.
Published on August 22, 2026, the paper presents Pandora’s Router as a model-selection approach that starts with cheap, noisy scores. It pays for better estimates when the expected value of that information exceeds its cost.
On the MATH, RAG, and EmbedLLM evaluation sets, the method had or tied for the lowest average combined inspection cost and routing regret, according to the available summary. To verify the scope of these results, consult the original paper and check its methodology, metric definitions, and evaluation details. If you use AI to study or apply the proposal, avoid entering sensitive organizational data or use an appropriate data protection policy.