Simple Yet Effective: An Information-Theoretic Approach to Multi-LLM Uncertainty Quantification
This addresses uncertainty quantification for high-stakes LLM applications, but it is incremental as it builds on existing ensemble and calibration techniques.
The paper tackled the problem of inconsistent LLM behavior by proposing MUSE, a method that aggregates diverse LLM outputs using Jensen-Shannon Divergence to improve uncertainty quantification, resulting in enhanced calibration and predictive performance on binary prediction tasks compared to baselines.
Large language models (LLMs) often behave inconsistently across inputs, indicating uncertainty and motivating the need for its quantification in high-stakes settings. Prior work on calibration and uncertainty quantification often focuses on individual models, overlooking the potential of model diversity. We hypothesize that LLMs make complementary predictions due to differences in training and the Zipfian nature of language, and that aggregating their outputs leads to more reliable uncertainty estimates. To leverage this, we propose MUSE (Multi-LLM Uncertainty via Subset Ensembles), a simple information-theoretic method that uses Jensen-Shannon Divergence to identify and aggregate well-calibrated subsets of LLMs. Experiments on binary prediction tasks demonstrate improved calibration and predictive performance compared to single-model and naïve ensemble baselines. In addition, we explore using MUSE as guided signals with chain-of-thought distillation to fine-tune LLMs for calibration. MUSE is available at:https://github.com/LARK-NLP-Lab/MUSE.