Bayesian Mixture of Experts For Large Language Models
This addresses the need for reliable uncertainty estimation in large language models for downstream decision-making, representing an incremental improvement over prior methods.
The paper tackles the problem of uncertainty estimation in fine-tuned large language models by proposing Bayesian Mixture of Experts (Bayesian-MoE), a post-hoc framework that improves expected calibration error and negative log-likelihood on benchmarks like Qwen1.5-MoE and DeepSeek-MoE.
We present Bayesian Mixture of Experts (Bayesian-MoE), a post-hoc uncertainty estimation framework for fine-tuned large language models (LLMs) based on Mixture-of-Experts architectures. Our method applies a structured Laplace approximation to the second linear layer of each expert, enabling calibrated uncertainty estimation without modifying the original training procedure or introducing new parameters. Unlike prior approaches, which apply Bayesian inference to added adapter modules, Bayesian-MoE directly targets the expert pathways already present in MoE models, leveraging their modular design for tractable block-wise posterior estimation. We use Kronecker-factored low-rank approximations to model curvature and derive scalable estimates of predictive uncertainty and marginal likelihood. Experiments on common-sense reasoning benchmarks with Qwen1.5-MoE and DeepSeek-MoE demonstrate that Bayesian-MoE improves both expected calibration error (ECE) and negative log-likelihood (NLL) over baselines, confirming its effectiveness for reliable downstream decision-making.