A hierarchical taxonomy for classifying hardness of inference tasks
International audience Exhibiting inferential capabilities is one of the major goals of many modern Natural Language Processing systems. However, if attempts have been made to define what textual inferences are, few seek to classify inference phenomena by difficulty. In this paper we propose a hiera...
Main Authors: | , |
---|---|
Other Authors: | , , , |
Format: | Conference Object |
Language: | English |
Published: |
HAL CCSD
2014
|
Subjects: | |
Online Access: | https://hal.archives-ouvertes.fr/hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982/document https://hal.archives-ouvertes.fr/hal-02281982/file/GleizeGrauLREC2014.pdf |
id |
ftccsdartic:oai:HAL:hal-02281982v1 |
---|---|
record_format |
openpolar |
spelling |
ftccsdartic:oai:HAL:hal-02281982v1 2023-05-15T16:50:05+02:00 A hierarchical taxonomy for classifying hardness of inference tasks Gleize, Martin Grau, Brigitte Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur (LIMSI) Université Paris Saclay (COmUE)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université - UFR d'Ingénierie (UFR 919) Sorbonne Université (SU)-Sorbonne Université (SU)-Université Paris-Saclay-Université Paris-Sud - Paris 11 (UP11) European Language Resources Association Reykjavik, Iceland 2014-05-26 https://hal.archives-ouvertes.fr/hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982/document https://hal.archives-ouvertes.fr/hal-02281982/file/GleizeGrauLREC2014.pdf en eng HAL CCSD hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982/document https://hal.archives-ouvertes.fr/hal-02281982/file/GleizeGrauLREC2014.pdf info:eu-repo/semantics/OpenAccess Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014) International Conference on Language Resources and Evaluation https://hal.archives-ouvertes.fr/hal-02281982 International Conference on Language Resources and Evaluation, European Language Resources Association, May 2014, Reykjavik, Iceland http://www.lrec-conf.org/proceedings/lrec2014/pdf/1168_Paper.pdf inference question answering textual entailment [INFO]Computer Science [cs] [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] info:eu-repo/semantics/conferenceObject Conference papers 2014 ftccsdartic 2021-12-19T01:49:40Z International audience Exhibiting inferential capabilities is one of the major goals of many modern Natural Language Processing systems. However, if attempts have been made to define what textual inferences are, few seek to classify inference phenomena by difficulty. In this paper we propose a hierarchical taxonomy for inferences, relatively to their hardness, and with corpus annotation and system design and evaluation in mind. Indeed, a fine-grained assessment of the difficulty of a task allows us to design more appropriate systems and to evaluate them only on what they are designed to handle. Each of seven classes is described and provided with examples from different tasks like question answering, textual entailment and coreference resolution. We then test the classes of our hierarchy on the specific task of question answering. Our annotation process of the testing data at the QA4MRE 2013 evaluation campaign reveals that it is possible to quantify the contrasts in types of difficulty on datasets of the same task. Conference Object Iceland Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe) |
institution |
Open Polar |
collection |
Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe) |
op_collection_id |
ftccsdartic |
language |
English |
topic |
inference question answering textual entailment [INFO]Computer Science [cs] [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] |
spellingShingle |
inference question answering textual entailment [INFO]Computer Science [cs] [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] Gleize, Martin Grau, Brigitte A hierarchical taxonomy for classifying hardness of inference tasks |
topic_facet |
inference question answering textual entailment [INFO]Computer Science [cs] [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] |
description |
International audience Exhibiting inferential capabilities is one of the major goals of many modern Natural Language Processing systems. However, if attempts have been made to define what textual inferences are, few seek to classify inference phenomena by difficulty. In this paper we propose a hierarchical taxonomy for inferences, relatively to their hardness, and with corpus annotation and system design and evaluation in mind. Indeed, a fine-grained assessment of the difficulty of a task allows us to design more appropriate systems and to evaluate them only on what they are designed to handle. Each of seven classes is described and provided with examples from different tasks like question answering, textual entailment and coreference resolution. We then test the classes of our hierarchy on the specific task of question answering. Our annotation process of the testing data at the QA4MRE 2013 evaluation campaign reveals that it is possible to quantify the contrasts in types of difficulty on datasets of the same task. |
author2 |
Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur (LIMSI) Université Paris Saclay (COmUE)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université - UFR d'Ingénierie (UFR 919) Sorbonne Université (SU)-Sorbonne Université (SU)-Université Paris-Saclay-Université Paris-Sud - Paris 11 (UP11) European Language Resources Association |
format |
Conference Object |
author |
Gleize, Martin Grau, Brigitte |
author_facet |
Gleize, Martin Grau, Brigitte |
author_sort |
Gleize, Martin |
title |
A hierarchical taxonomy for classifying hardness of inference tasks |
title_short |
A hierarchical taxonomy for classifying hardness of inference tasks |
title_full |
A hierarchical taxonomy for classifying hardness of inference tasks |
title_fullStr |
A hierarchical taxonomy for classifying hardness of inference tasks |
title_full_unstemmed |
A hierarchical taxonomy for classifying hardness of inference tasks |
title_sort |
hierarchical taxonomy for classifying hardness of inference tasks |
publisher |
HAL CCSD |
publishDate |
2014 |
url |
https://hal.archives-ouvertes.fr/hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982/document https://hal.archives-ouvertes.fr/hal-02281982/file/GleizeGrauLREC2014.pdf |
op_coverage |
Reykjavik, Iceland |
genre |
Iceland |
genre_facet |
Iceland |
op_source |
Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014) International Conference on Language Resources and Evaluation https://hal.archives-ouvertes.fr/hal-02281982 International Conference on Language Resources and Evaluation, European Language Resources Association, May 2014, Reykjavik, Iceland http://www.lrec-conf.org/proceedings/lrec2014/pdf/1168_Paper.pdf |
op_relation |
hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982 https://hal.archives-ouvertes.fr/hal-02281982/document https://hal.archives-ouvertes.fr/hal-02281982/file/GleizeGrauLREC2014.pdf |
op_rights |
info:eu-repo/semantics/OpenAccess |
_version_ |
1766040270482178048 |