COLING 2022main2 citations

Can We Guide a Multi-Hop Reasoning Language Model to Incrementally Learn at Each Single-Hop?

Jesus Lovon-Melgarejo, Jose G. Moreno, Romaric Besançon, Olivier Ferret, Lynda Tamine

Abstract

Despite the success of state-of-the-art pre-trained language models (PLMs) on a series of multi-hop reasoning tasks, they still suffer from their limited abilities to transfer learning from simple to complex tasks and vice-versa. We argue that one step forward to overcome this limitation is to better understand the behavioral trend of PLMs at each hop over the inference chain. Our critical underlying idea is to mimic human-style reasoning: we envision the multi-hop reasoning process as a sequence of explicit single-hop reasoning steps. To endow PLMs with incremental reasoning skills, we propose a set of inference strategies on relevant facts and distractors allowing us to build automatically generated training datasets. Using the SHINRA and ConceptNet resources jointly, we empirically show the effectiveness of our proposal on multiple-choice question answering and reading comprehension, with a relative improvement in terms of accuracy of 68.4% and 16.0% w.r.t. classic PLMs, respectively.

BibTeX
@inproceedings{lovon-melgarejo-etal-2022-guide,
    title = "Can We Guide a Multi-Hop Reasoning Language Model to Incrementally Learn at Each Single-Hop?",
    author = "Lovon-Melgarejo, Jesus  and
      Moreno, Jose G.  and
      Besan{\c{c}}on, Romaric  and
      Ferret, Olivier  and
      Tamine, Lynda",
    editor = "Calzolari, Nicoletta  and
      Huang, Chu-Ren  and
      Kim, Hansaem  and
      Pustejovsky, James  and
      Wanner, Leo  and
      Choi, Key-Sun  and
      Ryu, Pum-Mo  and
      Chen, Hsin-Hsi  and
      Donatelli, Lucia  and
      Ji, Heng  and
      Kurohashi, Sadao  and
      Paggio, Patrizia  and
      Xue, Nianwen  and
      Kim, Seokhwan  and
      Hahm, Younggyun  and
      He, Zhong  and
      Lee, Tony Kyungil  and
      Santus, Enrico  and
      Bond, Francis  and
      Na, Seung-Hoon",
    booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
    month = oct,
    year = "2022",
    address = "Gyeongju, Republic of Korea",
    publisher = "International Committee on Computational Linguistics",
    url = "https://aclanthology.org/2022.coling-1.125/",
    pages = "1455--1466"
}
Can We Guide a Multi-Hop Reasoning Language Model to Incrementally Learn at Each Single-Hop? · COLING 2022