Integrative Decoding: Improving Factuality via Implicit Self-consistency

Self-consistency-based approaches, which involve repeatedly sampling multiple outputs and selecting the most consistent one as the final response, prove to be remarkably effective in improving the factual accuracy of large language models. Nonetheless, existing methods usually have strict constraints on the task format, largely limiting their applicability. In this paper, we present Integrative Decoding (ID), to unlock the potential of self-consistency in open-ended generation tasks. ID operates by constructing a set of inputs, each prepended with a previously sampled response, and then processes them concurrently, with the next token being selected by aggregating of all their corresponding predictions at each decoding step. In essence, this simple approach implicitly incorporates self-consistency in the decoding objective. Extensive evaluation shows that ID consistently enhances factuality over a wide range of language models, with substantial improvements on the TruthfulQA (+11.2%), Biographies (+15.4%) and LongFact (+8.5%) benchmarks. The performance gains amplify progressively as the number of sampled responses increases, indicating the potential of ID to scale up with repeated sampling.

Authors

• Yi Cheng
• Xiao Liang
• Yeyun Gong
• Wen Xiao
• Song Wang
• Yuji Zhang 0002
• Wenjun Hou
• Kaishuai Xu
• Wenge Liu
• Wenjie Li 0002
• Jian Jiao 0007
• Qi Chen 0009
• Peng Cheng 0005
• Wayne Xiong

Keywords

• Large Language Models
• Hallucination
• Factuality
• Decoding