Papers by Shayan Hooshmand
Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics (2022.emnlp-main)
Copied to clipboard
Elisa Kreiss, Cynthia Bennett, Shayan Hooshmand, Eric Zelikman, Meredith Ringel Morris, Christopher Potts
| Challenge: | Existing referenceless metrics do not take context into account, whereas contextual information is highly valued by BLV users. |
| Approach: | They propose a contextual version of the referenceless metric CLIPScore which addresses the disconnect to the BLV data. |
| Outcome: | The proposed evaluation metrics are based on a proof-of-concept with blind and low vision (BLV) participants. |
CHoRaL: Collecting Humor Reaction Labels from Millions of Social Media Users (2021.emnlp-main)
Copied to clipboard
| Challenge: | Humor detection is difficult due to individualistic and cultural differences in humor perception . authors propose a framework to generate perceived humor labels on Facebook posts . |
| Approach: | They propose a framework to generate perceived humor labels on Facebook posts . they use the naturally available user reactions to these posts to generate the labels . |
| Outcome: | The proposed framework generates perceived humor labels on Facebook posts with no manual annotation needed. |