REBEL: Relation Extraction By End-to-end Language generation

TLDR

Extracting relation triplets from raw text is essential for tasks such as populating knowledge bases, yet existing pipelines are error‑prone and limited in relation types, while seq2seq models have proven effective in language generation and NLU tasks like entity linking. The paper introduces REBEL, an autoregressive seq2seq model that simplifies relation extraction by representing triplets as text sequences and supports over 200 relation types. REBEL adapts the BART architecture to generate relation triplets directly, enabling end‑to‑end extraction across many relation types. Fine‑tuning REBEL on multiple benchmarks achieves state‑of‑the‑art performance on most relation extraction and classification tasks.

Abstract

Extracting relation triplets from raw text is a crucial task in Information Extraction, enabling multiple applications such as populating or validating knowledge bases, factchecking, and other downstream tasks. However, it usually involves multiple-step pipelines that propagate errors or are limited to a small number of relation types. To overcome these issues, we propose the use of autoregressive seq2seq models. Such models have previously been shown to perform well not only in language generation, but also in NLU tasks such as Entity Linking, thanks to their framing as seq2seq tasks. In this paper, we show how Relation Extraction can be simplified by expressing triplets as a sequence of text and we present REBEL, a seq2seq model based on BART that performs end-to-end relation extraction for more than 200 different relation types. We show our model’s flexibility by fine-tuning it on an array of Relation Extraction and Relation Classification benchmarks, with it attaining state-of-the-art performance in most of them.