Papers by Phillip Lee
Athena 2.0: Contextualized Dialogue Management for an Alexa Prize SocialBot (2021.emnlp-demo)
Copied to clipboard
Juraj Juraska, Kevin Bowden, Lena Reed, Vrindavan Harrison, Wen Cui, Omkar Patil, Rishi Rajasekaran, Angela Ramirez, Cecilia Li, Eduardo Zamora, Phillip Lee, Jeshwanth Bheemanpally, Rohan Pandey, Adwait Ratnaparkhi, Marilyn Walker
| Challenge: | Athena 2.0 is a socialbot that has been a finalist in the last two Alexa Prize Grand Challenges. |
| Approach: | They describe Athena 2.0's dialogue management strategy and its performance in the Alexa Prize 20/21 competition. |
| Outcome: | The system is a finalist in the Alexa Prize 20/21 competition and will be shown on a live demo and recorded video recordings. |
Can vectors read minds better than experts? Comparing data augmentation strategies for the automated scoring of children’s mindreading ability (2021.acl-long)
Copied to clipboard
| Challenge: | In-domain experts are recruited to reannotate augmented samples and determine to what extent each strategy preserves the original rating. |
| Approach: | They implement 7 different data augmentation strategies for the task of automatic scoring of children’s ability to understand others’ thoughts, feelings, and desires. |
| Outcome: | The data augmentation strategies outperform task-agnostic augmentations and automatic augmentation systems perform worst on the MIND-CA corpus. |
“What is on your mind?” Automated Scoring of Mindreading in Childhood and Early Adolescence (2020.coling-main)
Copied to clipboard
Venelin Kovatchev, Phillip Smith, Mark Lee, Imogen Grumley Traynor, Irene Luque Aguilera, Rory Devine
| Challenge: | Existing studies show that children who excel at mindreading are more likely to be identified as popular by classmates and have reciprocated friendships. |
| Approach: | They propose to automate the scoring of mindreading ability in middle childhood and early adolescence using a new corpus of 11,311 question-answer pairs in English from 1,066 children aged from 7 to 14 . |
| Outcome: | The proposed scoring system is based on 11,311 question-answer pairs in English from 1,066 children aged from 7 to 14 . the results demonstrate the applicability of state-of-the-art NLP solutions to a new domain and task. |