TY - GEN

T1 - DetIE: Multilingual Open Information Extraction Inspired by Object Detection

AU - Vasilkovsky, Michael

AU - Alekseev, Anton

AU - Malykh, Valentin

AU - Shenbin, Ilya

AU - Tutubalina, Elena

AU - Salikhov, Dmitriy

AU - Stepnov, Mikhail

AU - Chertok, Andrey

AU - Nikolenko, Sergey

PY - 2022

Y1 - 2022

N2 - State of the art neural methods for open information extraction (OpenIE) usually extract triplets (or tuples) iteratively in an autoregressive or predicate-based manner in order not to produce duplicates. In this work, we propose a different approach to the problem that can be equally or more successful. Namely, we present a novel single-pass method for OpenIE inspired by object detection algorithms from computer vision. We use an order-agnostic loss based on bipartite matching that forces unique predictions and a Transformer-based encoder-only architecture for sequence labeling. The proposed approach is faster and shows superior or similar performance in comparison with state of the art models on standard benchmarks in terms of both quality metrics and inference time. Our model sets the new state of the art performance of 67.7% F1 on CaRB evaluated as OIE2016 while being 3.35x faster at inference than previous state of the art. We also evaluate the multilingual version of our model in the zero-shot setting for two languages and introduce a strategy for generating synthetic multilingual data to fine-tune the model for each specific language. In this setting, we show performance improvement 15% on multilingual Re-OIE2016, reaching 75% F1 for both Portuguese and Spanish languages. Code and models are available at https://github.com/sberbank-ai/DetIE.

AB - State of the art neural methods for open information extraction (OpenIE) usually extract triplets (or tuples) iteratively in an autoregressive or predicate-based manner in order not to produce duplicates. In this work, we propose a different approach to the problem that can be equally or more successful. Namely, we present a novel single-pass method for OpenIE inspired by object detection algorithms from computer vision. We use an order-agnostic loss based on bipartite matching that forces unique predictions and a Transformer-based encoder-only architecture for sequence labeling. The proposed approach is faster and shows superior or similar performance in comparison with state of the art models on standard benchmarks in terms of both quality metrics and inference time. Our model sets the new state of the art performance of 67.7% F1 on CaRB evaluated as OIE2016 while being 3.35x faster at inference than previous state of the art. We also evaluate the multilingual version of our model in the zero-shot setting for two languages and introduce a strategy for generating synthetic multilingual data to fine-tune the model for each specific language. In this setting, we show performance improvement 15% on multilingual Re-OIE2016, reaching 75% F1 for both Portuguese and Spanish languages. Code and models are available at https://github.com/sberbank-ai/DetIE.

KW - natural language processing

KW - open information extraction

KW - information extraction

KW - multilingual models

KW - deep learning

KW - object detection

UR - https://arxiv.org/search/cs?searchtype=author&query=Vasilkovsky%2C+M

UR - https://aaai.org/papers/11412-detie-multilingual-open-information-extraction-inspired-by-object-detection/

M3 - Conference contribution

SP - 11412

EP - 11420

BT - Proceedings of the 36th AAAI Conference on Artificial Intelligence, 36

T2 - 36th AAAI Conference on Artificial Intelligence

Y2 - 22 February 2022 through 1 March 2022

ER -