Relation extraction (RE) is the core NLP task of inferring semantic relationships between entities from text. Standard supervised RE techniques entail training modules to tag tokens comprising entity spans and then predict the relationship between them. Recent work has instead treated the problem as a \emph{sequence-to-sequence} task, linearizing relations between entities as target strings to be generated conditioned on the input. Here we push the limits of this approach, using larger language models (GPT-3 and Flan-T5 large) than considered in prior work and evaluating their performance on standard RE tasks under varying levels of supervision. We address issues inherent to evaluating generative approaches to RE by doing human evaluations, in lieu of relying on exact matching. Under this refined evaluation, we find that: (1) Few-shot prompting with GPT-3 achieves near SOTA performance, i.e., roughly equivalent to existing fully supervised models; (2) Flan-T5 is not as capable in the few-shot setting, but supervising and fine-tuning it with Chain-of-Thought (CoT) style explanations (generated via GPT-3) yields SOTA results. We release this model as a new baseline for RE tasks.

翻译：关系抽取（RE）是自然语言处理的核心任务，旨在从文本中推断实体间的语义关系。标准的监督式关系抽取技术包括训练模块来标记构成实体边界的词元，并预测它们之间的关系。近年来的研究则将该问题视为序列到序列任务，将实体间的关系线性化为目标字符串，并在输入条件下生成这些字符串。本研究推动这一方法的边界，使用比先前工作更大的语言模型（GPT-3和Flan-T5 large），评估它们在监督程度不同的标准关系抽取任务中的表现。我们通过人工评估（而非依赖精确匹配）来解决生成式关系抽取方法固有的评估问题。在这种精细化评估下，我们发现：（1）GPT-3的少样本提示性能接近当前最优水平，即与现有完全监督模型大致相当；（2）Flan-T5在少样本设置中能力稍弱，但通过GPT-3生成的思维链风格解释对其进行监督和微调，可获得当前最优结果。我们发布该模型作为关系抽取任务的新基线。