Papers by Raquel G. Alhama
Word Segmentation as Unsupervised Constituency Parsing (2022.acl-long)
Copied to clipboard
| Challenge: | Existing theories of word identification from continuous inputs are based on statistical cues, such as Bayesian inference and normative statistics. |
| Approach: | They propose a model which allows for a process isomorphic to unsupervised constituency parsing and which can reproduce human behavior in word identification experiments. |
| Outcome: | The proposed model reproduces human behavior in word identification experiments, suggesting it is viable to study word identification and its relation to syntactic processing. |