n this paper, we define and evaluate a methodology for extracting history-dependent spatial questions from visual dialogues. We say that a question is history-dependent if it requires (parts of) its dialogue history to be interpreted. We argue that some kinds of visual questions define a context upon which a follow-up spatial question relies. We call the question that restricts the context: trigger, and we call the spatial question that requires the trigger question to be answered: zoomer. We automatically extract different trigger and zoomer pairs based on the visual property that the questions rely on (e.g. color, number). We manually annotate the automatically extracted trigger and zoomer pairs to verify which zoomers require their trigger. We implement a simple baseline architecture based on a SOTA multimodal encoder. Our results reveal that there is much room for improvement for answering history-dependent questions.
Visually Grounded Follow-up Questions: a Dataset of Spatial Questions Which Require Dialogue History / Dong, Tianai; Testoni, Alberto; Benotti, Luciana; Bernardi, Raffaella. - ELETTRONICO. - (2021), pp. 22-31. (Intervento presentato al convegno Splurobonlp tenutosi a online nel 5-6 August 2021) [10.18653/v1/2021.splurobonlp-1.3].
Visually Grounded Follow-up Questions: a Dataset of Spatial Questions Which Require Dialogue History
Testoni, Alberto;Bernardi, Raffaella
2021-01-01
Abstract
n this paper, we define and evaluate a methodology for extracting history-dependent spatial questions from visual dialogues. We say that a question is history-dependent if it requires (parts of) its dialogue history to be interpreted. We argue that some kinds of visual questions define a context upon which a follow-up spatial question relies. We call the question that restricts the context: trigger, and we call the spatial question that requires the trigger question to be answered: zoomer. We automatically extract different trigger and zoomer pairs based on the visual property that the questions rely on (e.g. color, number). We manually annotate the automatically extracted trigger and zoomer pairs to verify which zoomers require their trigger. We implement a simple baseline architecture based on a SOTA multimodal encoder. Our results reveal that there is much room for improvement for answering history-dependent questions.File | Dimensione | Formato | |
---|---|---|---|
2021.splurobonlp-1.3.pdf
accesso aperto
Descrizione: articolo principale
Tipologia:
Versione editoriale (Publisher’s layout)
Licenza:
Creative commons
Dimensione
5.7 MB
Formato
Adobe PDF
|
5.7 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione