Search papers, labs, and topics across Lattice.
This paper introduces a novel routing mechanism for mixtures of low-rank adaptation (LoRA) experts, termed VI-MoLE, which leverages certified value-of-information to optimize expert activation based on the marginal risk reduction per unit cost. By distinguishing between present ambiguity and recoverable risk, VI-MoLE improves decision-making in uncertain scenarios, ensuring that only the most informative experts are activated. The results demonstrate that VI-MoLE outperforms traditional uncertainty-based routing methods in terms of accuracy and efficiency across various evaluation metrics, including distribution shift and tail latency.
Uncertainty alone can mislead expert activation; VI-MoLE redefines routing by focusing on certified value-of-information, leading to more informed and efficient model decisions.
Mixtures of low-rank adaptation experts increase parameter-efficient capacity by routing each input through a subset of adapters. Recent dynamic routers activate more experts when the router or prediction is uncertain. This rule silently equates uncertainty with useful additional computation: an uncertain example may contain complementary, unqueried expert evidence, but it may instead remain ambiguous after every expert agrees. We formulate routing as certified value-of-information allocation. VI-MoLE learns the counterfactual risk remaining after each expert prefix, converts these predictions into simultaneous upper-risk certificates on held-out calibration data, and spends a global adapter budget on the token--layer action with the largest certified marginal risk reduction per unit cost. A terminal certificate then decides whether to answer or abstain. Unlike an uncertainty gate, this procedure distinguishes present ambiguity from recoverable and residual risk. We prove simultaneous certificate validity, optimal greedy allocation under diminishing certified gains, and allocation regret under value-estimation error. The evaluation protocol tests matched-compute accuracy, certificate coverage, risk--coverage, distribution shift, and tail latency against fixed and dynamic MoE-LoRA routers.
UW