Timezone: »
Knowledge-intensive language tasks require NLP systems to both provide the correct answer and retrieve supporting evidence for it in a given corpus. Autoregressive language models are emerging as the de-facto standard for generating answers, with newer and more powerful systems emerging at an astonishing pace. In this paper we argue that all this (and future) progress can be directly applied to the retrieval problem with minimal intervention to the models' architecture. Previous work has explored ways to partition the search space into hierarchical structures and retrieve documents by autoregressively generating their unique identifier. In this work we propose an alternative that doesn't force any structure in the search space: using all ngrams in a passage as its possible identifiers. This setup allows us to use an autoregressive model to generate and score distinctive ngrams, that are then mapped to full passages through an efficient data structure. Empirically, we show this not only outperforms prior autoregressive approaches but also leads to an average improvement of at least 10 points over more established retrieval solutions for passage-level retrieval on the KILT benchmark, establishing new state-of-the-art downstream performance on some datasets, while using a considerably lighter memory footprint than competing systems. Code available in the supplementary materials. Pre-trained models will be made available.
Author Information
Michele Bevilacqua (Samaya AI)
Giuseppe Ottaviano (Facebook)
Patrick Lewis (FAIR)
Scott Yih (Meta AI - FAIR)
Sebastian Riedel (DeepMind / UCL)
Fabio Petroni (Facebook AI Research)
More from the Same Authors
-
2023 Poster: Improving Language Plasticity via Pretraining with Active Forgetting »
Yihong Chen · Mikel Artetxe · Kelly Marchisio · Roberta Raileanu · David Adelani · Pontus Lars Erik Saito Stenetorp · Sebastian Riedel -
2022 Spotlight: BiT: Robustly Binarized Multi-distilled Transformer »
Zechun Liu · Barlas Oguz · Aasish Pappu · Lin Xiao · Scott Yih · Meng Li · Raghuraman Krishnamoorthi · Yashar Mehdad -
2022 Poster: BiT: Robustly Binarized Multi-distilled Transformer »
Zechun Liu · Barlas Oguz · Aasish Pappu · Lin Xiao · Scott Yih · Meng Li · Raghuraman Krishnamoorthi · Yashar Mehdad -
2022 Poster: ReFactor GNNs: Revisiting Factorisation-based Models from a Message-Passing Perspective »
Yihong Chen · Pushkar Mishra · Luca Franceschi · Pasquale Minervini · Pontus Lars Erik Saito Stenetorp · Sebastian Riedel -
2017 Workshop: Workshop on Prioritising Online Content »
John Shawe-Taylor · Massimiliano Pontil · Nicolò Cesa-Bianchi · Emine Yilmaz · Chris Watkins · Sebastian Riedel · Marko Grobelnik -
2017 : Reading and Reasoning with Neural Program Interpreters »
Sebastian Riedel -
2017 Poster: End-to-End Differentiable Proving »
Tim Rocktäschel · Sebastian Riedel -
2017 Oral: End-to-end Differentiable Proving »
Tim Rocktäschel · Sebastian Riedel -
2016 Workshop: Neural Abstract Machines & Program Induction »
Matko Bošnjak · Nando de Freitas · Tejas Kulkarni · Arvind Neelakantan · Scott E Reed · Sebastian Riedel · Tim Rocktäschel -
2014 Workshop: 3rd NIPS Workshop on Probabilistic Programming »
Daniel Roy · Josh Tenenbaum · Thomas Dietterich · Stuart J Russell · YI WU · Ulrik R Beierholm · Alp Kucukelbir · Zenna Tavares · Yura Perov · Daniel Lee · Brian Ruttenberg · Sameer Singh · Michael Hughes · Marco Gaboardi · Alexey Radul · Vikash Mansinghka · Frank Wood · Sebastian Riedel · Prakash Panangaden -
2014 Workshop: 4th Workshop on Automated Knowledge Base Construction (AKBC) »
Sameer Singh · Fabian M Suchanek · Sebastian Riedel · Partha Pratim Talukdar · Kevin Murphy · Christopher Ré · William Cohen · Tom Mitchell · Andrew McCallum · Jason E Weston · Ramanathan Guha · Boyan Onyshkevych · Hoifung Poon · Oren Etzioni · Ari Kobren · Arvind Neelakantan · Peter Clark -
2014 Demonstration: A Visual and Interactive IDE for Probabilistic Programming »
Sameer Singh · Luke Hewitt · Tim Rocktäschel · Sebastian Riedel