Papers by Casey Kennington
Conceptual Pacts for Reference Resolution Using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing large language models can be fine-tuned offline but are large and resource-intensive. |
| Approach: | They propose to use a simple reference resolver to simulate a conceptual pact process over time with different conversation pairs. |
| Outcome: | The proposed model performs better than a pre-trained model with exhaustive retraining after each prediction, while being more transparent, faster and less resource-intensive. |
Incorporating Word-level Phonemic Decoding into Readability Assessment (2024.lrec-main)
Copied to clipboard
| Challenge: | a recent study suggests that automatic readability assessment is not able to provide interpretability for teachers and educators. |
| Approach: | They propose to incorporate phonetic and orthographic features into automatic readability assessment by handcrafted feature sets. |
| Outcome: | a new feature set shows comparable performance to larger feature sets on grade-level classification in english . authors say the model improves on multiple readability datasets but lacks interpretability . |
HADREB: Human Appraisals and (English) Descriptions of Robot Emotional Behaviors (2022.lrec-1)
Copied to clipboard
| Challenge: | HADREB datasets explore how humans perceive robot emotional states . emotions are a fundamental part of the human language system and are used as scaffolding for language learning . |
| Approach: | They present a dataset of human appraisals and English descriptions of robot emotional behaviors . they use mistyrobotics mist and digital dream labs cozmo robots to analyze the data . |
| Outcome: | The proposed dataset examines how humans perceive robot emotional states and how they relate to human language. |
KidSpell: A Child-Oriented, Rule-Based, Phonetic Spellchecker (2020.lrec-1)
Copied to clipboard
Brody Downs, Oghenemaro Anuyah, Aprajita Shukla, Jerry Alan Fails, Sole Pera, Katherine Wright, Casey Kennington
| Challenge: | Existing spellcheckers are tuned to the needs of adults and are unsatisfactory for children due to their varying cognitive capabilities. |
| Approach: | They propose a model that maps misspelled words and spelling suggestions to their phonetic keys and a selection process that prioritizes candidate spelling suggestions that closely align with the misspelled word. |
| Outcome: | The proposed model outperforms existing spellcheckers in a number of offline experiments using existing and novel datasets. |
SafetyALFRED: Evaluating Safety-Conscious Planning of Vision Language Models (2026.findings-acl)
Copied to clipboard
Josue Torres-Fonseca, Naihao Deng, Yinpei Dai, Shane Storks, Yichi Zhang, Rada Mihalcea, Casey Kennington, Joyce Chai
| Challenge: | Existing safety evaluations focus on hazard recognition through disembodied question answering (QA) settings, but lack a critical gap in evaluating an agent. |
| Approach: | They evaluate multimodal large language models with six categories of kitchen hazards . they propose a safety-based approach that prioritizes multi-step corrective actions . |
| Outcome: | The proposed model can recognize hazards in QA settings, but average mitigation success rates are low . the proposed model is based on the embodied agent benchmark ALFRED . |
Evaluating and Improving Child-Directed Automatic Speech Recognition (2020.lrec-1)
Copied to clipboard
| Challenge: | a recent study shows that adult speech recognition systems are lagging behind child models due to the fact that children's vocal tracts are smaller than adults . |
| Approach: | They evaluate a model that trains on adult data and apply additional tuning to varied amounts of child speech data to improve child-directed speech recognition. |
| Outcome: | The proposed model improves over baseline models using child data and small amounts of child audio data. |