Visa enkel post

dc.contributor.authorAruqi, Ali
dc.date.accessioned2021-11-12T07:35:44Z
dc.date.available2021-11-12T07:35:44Z
dc.date.issued2021-11-12
dc.identifier.urihttp://hdl.handle.net/2077/70001
dc.description.abstractEmbodied question answering is the task of asking a robot about objects in a 3D environment. The robot has to navigate the environment, find the entities in question, and then stop to answer the question. The answering system consists of navigation and visual-question-answering components. The agent is trained on a synthetic data-set of question-answers and navigational paths called EQA-MP3D. Each question in the data-set is an executable function that could be run in the environment to yield an answer. EQA-MP3D includes only two types of questions, color and location questions. The type of questions asked could be considered unnatural, and we observe that the question-answers contain biases. Our work extends the data-set by automatically generating size and spatial questions. We generate a total of 19 207 question-answers for training and 3 186 question-answers for validation. Our data extension is intended to train the system to answer more question types and enhance the system’s overall ability to perform the task.sv
dc.language.isoengsv
dc.subjectEmbodied Question Answeringsv
dc.subjectQuestion Generationsv
dc.subjectSpatial Relationssv
dc.subjectSynthetic Data-setssv
dc.subjectMulti-Modalitysv
dc.titleEMBODIED QUESTION ANSWERING IN ROBOTIC ENVIRONMENT Automatic generation of a synthetic question-answer data-setsv
dc.typeText
dc.setspec.uppsokHumanitiesTheology
dc.type.uppsokH2
dc.contributor.departmentUniversity of Gothenburg / Department of Philosophy,Lingustics and Theory of Scienceeng
dc.contributor.departmentGöteborgs universitet / Institutionen för filosofi, lingvistik och vetenskapsteoriswe
dc.type.degreeStudent essay


Filer under denna titel

Thumbnail

Dokumentet tillhör följande samling(ar)

Visa enkel post