Teach the Rules, Provide the Facts: Targeted Relational-knowledge Enhancement for Textual Inference

Ohad Rozen, Shmuel Amar, Vered Shwartz, Ido Dagan

פרסום מחקרי: פרק בספר / בדוח / בכנספרסום בספר כנסביקורת עמיתים

תקציר

We present InferBert, a method to enhance transformer-based inference models with relevant relational knowledge. Our approach facilitates learning generic inference patterns requiring relational knowledge (e.g. inferences related to hypernymy) during training, while injecting on-demand the relevant relational facts (e.g. pangolin is an animal) at test time. We apply InferBERT to the NLI task over a diverse set of inference types (hypernymy, location, color, and country of origin), for which we collected challenge datasets. In this setting, InferBert succeeds to learn general inference patterns, from a relatively small number of training instances, while not hurting performance on the original NLI data and substantially outperforming prior knowledge enhancement models on the challenge data. It further applies its inferences successfully at test time to previously unobserved entities. InferBert is computationally more efficient than most prior methods, in terms of number of parameters, memory consumption and training time.

שפה מקוריתאנגלית
כותר פרסום המארח*SEM 2021 - 10th Conference on Lexical and Computational Semantics, Proceedings of the Conference
עורכיםLun-Wei Ku, Vivi Nastase, Ivan Vulic
מוציא לאורAssociation for Computational Linguistics (ACL)
עמודים89-98
מספר עמודים10
מסת"ב (אלקטרוני)9781954085770
סטטוס פרסוםפורסם - 2021
אירוע10th Conference on Lexical and Computational Semantics, *SEM 2021 - Virtual, Bangkok, תאילנד
משך הזמן: 5 אוג׳ 20216 אוג׳ 2021

סדרות פרסומים

שם*SEM 2021 - 10th Conference on Lexical and Computational Semantics, Proceedings of the Conference

כנס

כנס10th Conference on Lexical and Computational Semantics, *SEM 2021
מדינה/אזורתאילנד
עירVirtual, Bangkok
תקופה5/08/216/08/21

ASJC Scopus subject areas

  • ???subjectarea.asjc.1700.1703???
  • ???subjectarea.asjc.1700.1705???
  • ???subjectarea.asjc.1700.1706???
  • ???subjectarea.asjc.1700.1710???
  • ???subjectarea.asjc.2600.2614???

טביעת אצבע

להלן מוצגים תחומי המחקר של הפרסום 'Teach the Rules, Provide the Facts: Targeted Relational-knowledge Enhancement for Textual Inference'. יחד הם יוצרים טביעת אצבע ייחודית.

פורמט ציטוט ביבליוגרפי