Can Visual Dialogue Models Do Scorekeeping? Exploring How Dialogue Representations Incrementally Encode Shared Knowledge
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913705236627456 |
|---|---|
| author | Madureira, Brielen Schlangen, David |
| author_facet | Madureira, Brielen Schlangen, David |
| contents | Cognitively plausible visual dialogue models should keep a mental scoreboard of shared established facts in the dialogue context. We propose a theory-based evaluation method for investigating to what degree models pretrained on the VisDial dataset incrementally build representations that appropriately do scorekeeping. Our conclusion is that the ability to make the distinction between shared and privately known statements along the dialogue is moderately present in the analysed models, but not always incrementally consistent, which may partially be due to the limited need for grounding interactions in the original task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2204_06970 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Can Visual Dialogue Models Do Scorekeeping? Exploring How Dialogue Representations Incrementally Encode Shared Knowledge Madureira, Brielen Schlangen, David Computation and Language Cognitively plausible visual dialogue models should keep a mental scoreboard of shared established facts in the dialogue context. We propose a theory-based evaluation method for investigating to what degree models pretrained on the VisDial dataset incrementally build representations that appropriately do scorekeeping. Our conclusion is that the ability to make the distinction between shared and privately known statements along the dialogue is moderately present in the analysed models, but not always incrementally consistent, which may partially be due to the limited need for grounding interactions in the original task. |
| title | Can Visual Dialogue Models Do Scorekeeping? Exploring How Dialogue Representations Incrementally Encode Shared Knowledge |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2204.06970 |