MetaQA: Combining Expert Agents for Multi-Skill Question Answering
This work addresses the need for adaptable and efficient multi-skill question answering systems, though it is incremental as it builds on existing multi-agent and multi-dataset methods.
The paper tackles the problem of multi-skill question answering by combining expert agents with a flexible architecture that uses questions, answer predictions, and confidence scores to select the best answer, resulting in outperforming previous multi-agent and multi-dataset approaches in both in-domain and out-of-domain scenarios.
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 https://github.com/UKPLab/MetaQA.