References & Citations
Computer Science > Computation and Language
Title: MetaQA: Combining Expert Agents for Multi-Skill Question Answering
(Submitted on 3 Dec 2021 (v1), last revised 7 Feb 2023 (this version, v4))
Abstract: The recent explosion of question answering (QA) datasets and models has increased the interest in the generalization of models across multiple domains and formats by either training on multiple datasets or by combining multiple models. Despite the promising results of multi-dataset models, some domains or QA formats may require specific architectures, and thus the adaptability of these models might be limited. In addition, current approaches for combining models disregard cues such as question-answer compatibility. In this work, we propose to combine expert agents with a novel, flexible, and training-efficient architecture that considers questions, answer predictions, and answer-prediction confidence scores to select the best answer among a list of answer candidates. Through quantitative and qualitative experiments we show that our model i) creates a collaboration between agents that outperforms previous multi-agent and multi-dataset approaches in both in-domain and out-of-domain scenarios, ii) is highly data-efficient to train, and iii) can be adapted to any QA format. We release our code and a dataset of answer predictions from expert agents for 16 QA datasets to foster future developments of multi-agent systems on this https URL
Submission history
From: Haritz Puerto [view email][v1] Fri, 3 Dec 2021 14:05:52 GMT (6556kb,D)
[v2] Tue, 14 Dec 2021 12:48:57 GMT (6534kb,D)
[v3] Sun, 22 Jan 2023 12:54:40 GMT (6824kb,D)
[v4] Tue, 7 Feb 2023 03:15:32 GMT (6825kb,D)
Link back to: arXiv, form interface, contact.