The Provenance Paradox in Multi-Agent LLM Routing: Delegation Contracts and Attested Identity in LDP
This addresses a critical trust issue in multi-agent LLM routing for AI systems, offering a solution to prevent performance degradation from dishonest delegates.
The paper tackles the problem of multi-agent LLM systems selecting poor delegates due to inflated self-reported quality scores, showing that quality-based routing performs worse than random selection (e.g., 0.55 vs. 0.68 in simulations), and proposes extensions to LDP that achieve near-optimal performance (d = 9.51).
Multi-agent LLM systems delegate tasks across trust boundaries, but current protocols do not govern delegation under unverifiable quality claims. We show that when delegates can inflate self-reported quality scores, quality-based routing produces a provenance paradox: it systematically selects the worst delegates, performing worse than random. We extend the LLM Delegate Protocol (LDP) with delegation contracts that bound authority through explicit objectives, budgets, and failure policies; a claimed-vs-attested identity model that distinguishes self-reported from verified quality; and typed failure semantics enabling automated recovery. In controlled experiments with 10 simulated delegates and validated with real Claude models, routing by self-claimed quality scores performs worse than random selection (simulated: 0.55 vs. 0.68; real models: 8.90 vs. 9.30), while attested routing achieves near-optimal performance (d = 9.51, p < 0.001). Sensitivity analysis across 36 configurations confirms the paradox emerges reliably when dishonest delegates are present. All extensions are backward-compatible with sub-microsecond validation overhead.