Scribd Logo
Upload

Welcome to Scribd!

  • Poster On Explanation Regeneration Via Multi-Hop ILP Inference Over Knowledge Base

    Uploaded by

    Aayushee Gupta
    11 views1 page
    This poster describes a system for generating simple sentence explanations for complex scientific phenomena related to question answers. The system uses multiple components: a BERT model to initially rank facts from a knowledge base, an integer linear program to rescore the facts based on alignments between question, answer, and facts, and a regression model to further re-rank the facts into an explanatory chain. On the shared task, the system achieved a mean average precision of 0.36592. While the addition of an inter-fact chain alignment feature did not improve performance, the use of an ILP to link questions, answers, and facts was able to better infer explanatory chains compared to ranking facts individually.

    Original Description:

    Original Title

    Poster on Explanation Regeneration via Multi-Hop ILP Inference over Knowledge Base

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    This poster describes a system for generating simple sentence explanations for complex scientific phenomena related to question answers. The system uses multiple components: a BERT model to initially rank facts from a knowledge base, an integer linear program to rescore the facts based on alignments between question, answer, and facts, and a regression model to further re-rank the facts into an explanatory chain. On the shared task, the system achieved a mean average precision of 0.36592. While the addition of an inter-fact chain alignment feature did not improve performance, the use of an ILP to link questions, answers, and facts was able to better infer explanatory chains compared to ranking facts individually.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    11 views1 page

    Poster On Explanation Regeneration Via Multi-Hop ILP Inference Over Knowledge Base

    Uploaded by

    Aayushee Gupta
    This poster describes a system for generating simple sentence explanations for complex scientific phenomena related to question answers. The system uses multiple components: a BERT model to initially rank facts from a knowledge base, an integer linear program to rescore the facts based on alignments between question, answer, and facts, and a regression model to further re-rank the facts into an explanatory chain. On the shared task, the system achieved a mean average precision of 0.36592. While the addition of an inter-fact chain alignment feature did not improve performance, the use of an ILP to link questions, answers, and facts was able to better infer explanatory chains compared to ranking facts individually.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 1

    Aayushee Gupta, G.

    Srinivasaraghavan
    International Institute of Information Technology Bangalore

    Abstract System Description Evaluation Results


    Explanation regeneration is the task of
    generating simple sentence explanations for
    complex scientific phenomena related
    question answers [1].

    This poster describes our system submission


    to the Textgraphs 2020 workshop shared task
    which is based on multiple components: a
    BERT baseline ranking, an Integer Linear
    Program (ILP) based re-scoring, and a
    regression model for re-ranking the
    explanation facts. Our system achieved a
    Mean Average Precision score of 0.36592.

    • Generate a baseline ranking of individual facts in KB by fine-tuning BERT model[2] and get top-K ranked
    facts with respect to each question and its correct answer
    Task Description • Use the fact chain of top-K ranked facts and create an Integer Linear Program (ILP) with variables and Conclusion
    constraints consisting of a graph of nodes and edges created from constituents of question, answer and
    A multi-hop inferencing task of regenerating the fact chain. The ILP maximizes the graph[3] that has maximum alignment between the edges for re- Considering the multi-hop nature of the task,
    explanations wherein the gold explanation is scoring each fact in the chain in order to gather the relevant subgraph of correct chain of facts that can instead of ranking each fact from the KB
    formed by chaining together individual facts explain the answer: individually, we devised an ILP that can infer
    from a Knowledge Base (KB) ordered such that links between question, answer and KB facts to
    they form correct reasoning behind the score and rank a chain of explanation facts with
    answer to a question. • Different Wordnet- based alignment scores used for computing phrasal entailment (edge) scores between the regression model.
    constituent nodes of question, answer and fact chain include: Question Fact Chain Alignments (QFA), Fact
    Chain Answer Alignments (FAA), Intra-Fact Chain Alignments (IFA), Inter-Fact Chain Alignments (IFA2). The addition of Inter-Fact Chain Alignment
    • Generate combined scores from a regression model that takes alignment scores from the ILP model for feature did not improve the MAP score
    each QA pair as input features and produce a final sorted list of scores to re-rank facts in a chain leading probably because of the semantic drift among
    to generation of the final explanation for each question: facts in an explanation indicating requirement
    of better ILP parameter tuning and phrasal
    entailment scoring methods.

    Literature cited Acknowledgments Further information


    [1] Zhengnan Xie, Sebastian Thiem, Jaycie Martin, Elizabeth Wainwright, Steven Marmorstein, and
    Peter Jansen. 2020. Worldtree v2: A corpus of science-domain structured explanations and
    We thank the Textgraphs 2020 shared task organizers for arranging Our code is available at: https://github.com/aayushee/Textgraphs
    inference patterns supporting multi-hop inference. In Proceedings of The 12th Language such an interesting quest that gave us a great learning experience For any clarification, please contact: aayushee.gupta@iiitb.org
    Resources and Evaluation Conference, pages 5456–5473. and the anonymous paper reviewers for their helpful comments.
    [2] Rajarshi Das, Ameya Godbole, Manzil Zaheer, Shehzaad Dhuliawala, and Andrew McCallum.
    2019. Chainsof-reasoning at textgraphs 2019 shared task: Reasoning over chains of facts for
    explainable multi-hop inference. In Proceedings of the Thirteenth Workshop on Graph-Based
    Methods for Natural Language Processing (TextGraphs 13), pages 101–117.
    [3] Daniel Khashabi, Tushar Khot, Ashish Sabharwal, and Dan Roth. 2019. Question answering as
    global reasoning over semantic abstractions. arXiv preprint arXiv:1906.03672.

    You might also like