ROUTEAUDIT: Interaction-Aware Identification for Budgeted Multi-Verifier Routing

2026-10-02

AI Summary

This study addresses the evaluation bias and attribution challenges in multi-verifier routing caused by shifting contract conditions, formulating routing as a contract condition identification problem. Methodologically, it introduces core mechanisms including contract lattices, policy-agnostic response bands, and request-level bounds. By integrating causal inference, adaptive policy learning, and RLVR, the approach achieves precise attribution through matching path comparisons and generates verifiable certificates. Experimental results demonstrate that the proposed method effectively decouples policy gains from verifier set discrepancies, reducing attribution error to 0.0011 and significantly enhancing the reliability of system evaluation.

Abstract

Adaptive multi-verifier systems are commonly compared through endpoint quality-cost gaps, even when the verifier catalog, availability, accounting, information filtration, or scorer changes with the policy. We formulate verifier routing as a contract-conditioned identification problem. The contract records request support, verifier catalog, realized availability, resource accounting, online filtration, and post-trace scoring; a matched route contrast changes only the policy coordinate. ROUTEAUDIT adds three measurable objects to this contract. A contract lattice averages coordinate increments over every admissible bridge order and reports the resulting attribution together with its path sensitivity. A policy-independent response tape identifies paired sequential contrasts when adaptive policies reveal different observations. For incomplete matching, request-level bounds use whichever potential outcome remains observed and give a sharp finite-population interval. The protocol commits paid observations and ledger events before the oracle join and returns an attribution certificate for each comparison. On two held-out raw-tail caches, matched static SF+SA equals the cascade, assigning the apparent gains of 0.1797 and 0.1250 over full static to the verifier-set edge. On 1,319 held-out task requests, the learned and RLVR studies report quality 0.9522 and 0.9553 versus 0.9484 for matched static; the RLVR-static paired difference is +0.0068 with a request-paired interval $[0.0015,0.0122]$ and a training-seed-by-request hierarchical interval $[0.0006,0.0131]$. Controlled attribution recovery yields route mean absolute error 0.0011 and endpoint reconstruction error 0.0004. Factorial, bridge-order, and stochastic-provider studies evaluate the certificate interface; RLVR supplies a learned-policy stress test under the same identification contract.

Problem

• multi-verifier routing
• budgeted routing
• causal identification
• fair evaluation
• policy attribution

Innovation

• Multi-Verifier Routing
• Contract-Conditioned Identification
• Attribution Certificate
• Causal Inference
• Response Tape