Boosted Prompt Ensembles for Large Language Models
This addresses the challenge of enhancing reasoning in large language models for tasks like math and logic, offering an incremental improvement over existing prompting techniques.
The paper tackles the problem of improving language model reasoning performance without additional training by proposing a boosted prompt ensembling method that selects hard examples stepwise to construct few-shot prompts, resulting in outperformance over single-prompt and bagged ensembles on datasets like GSM8k and AQuA.
Methods such as chain-of-thought prompting and self-consistency have pushed the frontier of language model reasoning performance with no additional training. To further improve performance, we propose a prompt ensembling method for large language models, which uses a small dataset to construct a set of few shot prompts that together comprise a ``boosted prompt ensemble''. The few shot examples for each prompt are chosen in a stepwise fashion to be ``hard'' examples on which the previous step's ensemble is uncertain. We show that this outperforms single-prompt output-space ensembles and bagged prompt-space ensembles on the GSM8k and AQuA datasets, among others. We propose both train-time and test-time versions of boosted prompting that use different levels of available annotation and conduct a detailed empirical study of our algorithm.