SimpleStrat: Diversifying Language Model Generation with Stratification

TL;DR

SimpleStrat is a novel method that enhances diversity in language model outputs by stratifying the response space, outperforming temperature-based methods in quality and diversity metrics.

Contribution

We introduce SimpleStrat, a stratification-based approach for diversifying language model responses, and propose CoverageQA to evaluate diversity effectively.

Findings

SimpleStrat achieves higher recall than GPT-4o.

SimpleStrat reduces KL Divergence by 0.36 on average.

Temperature increase lowers response quality, contrary to prior belief.

Abstract

Generating diverse responses from large language models (LLMs) is crucial for applications such as planning/search and synthetic data generation, where diversity provides distinct answers across generations. Prior approaches rely on increasing temperature to increase diversity. However, contrary to popular belief, we show not only does this approach produce lower quality individual generations as temperature increases, but it depends on model's next-token probabilities being similar to the true distribution of answers. We propose SimpleStrat, an alternative approach that uses the language model itself to partition the space into strata. At inference, a random stratum is selected and a sample drawn from within the strata. To measure diversity, we introduce CoverageQA, a dataset of underspecified questions with multiple equally plausible answers, and assess diversity by measuring KL Divergence between the output distribution and uniform distribution over valid ground truth answers. As computing probability per response/solution for proprietary models is infeasible, we measure recall on ground truth solutions. Our evaluation show using SimpleStrat achieves higher recall by 0.05 compared to GPT-4o and 0.36 average reduction in KL Divergence compared to Llama 3.