Don't Throw Away Your Beams: Improving Consistency-based Uncertainties in LLMs via Beam Search
This work addresses uncertainty quantification for users of large language models in short-form QA, offering an incremental improvement over existing methods.
The paper tackled the problem of high variance and duplicate generations in consistency-based uncertainty quantification for large language models by introducing beam search as an alternative to multinomial sampling, resulting in state-of-the-art performance on six QA datasets with reduced variance.
Consistency-based methods have emerged as an effective approach to uncertainty quantification (UQ) in large language models. These methods typically rely on several generations obtained via multinomial sampling, measuring their agreement level. However, in short-form QA, multinomial sampling is prone to producing duplicates due to peaked distributions, and its stochasticity introduces considerable variance in uncertainty estimates across runs. We introduce a new family of methods that employ beam search to generate candidates for consistency-based UQ, yielding improved performance and reduced variance compared to multinomial sampling. We also provide a theoretical lower bound on the beam set probability mass under which beam search achieves a smaller error than multinomial sampling. We empirically evaluate our approach on six QA datasets and find that its consistent improvements over multinomial sampling lead to state-of-the-art UQ performance.