DPAEG: A Dependency Parse-Based Adversarial Examples Generation Method for Intelligent Q&A Robots

<table class="table-group" id="tab4"><tr><td><table class="table"><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr class="thead"><td class="align_left">Target Q&amp;A robot</td><td class="align_center">Original questions (%)</td><td class="align_center">Restated questions (%)</td></tr><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr><td class="align_left">DrQA</td><td class="align_center">78.21</td><td class="align_center">76.05</td></tr><tr><td class="align_left">Google Assistant</td><td class="align_center">49.48</td><td class="align_center">48.73</td></tr><tr class="table-tr"><td colspan="3"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Success rates of the adversarial questions generated from the original questions and from the restated questions on the two target Q&amp;A robots.</div>

Security and Communication Networks

tab4

Table 4

Table 4: DPAEG: A Dependency Parse-Based Adversarial Examples Generation Method for Intelligent Q&A Robots 