The evaluation data used in Sadeghi, S., Scheutz, M. (2018). Sensitivity to Input Order: Evaluation of an Incremental and Memory-Limited Bayesian Cross-Situational Word Learning Model. In Proceedings of the 27th International Conference on Computational Linguistics (COLING 2018).
- Each folder contains the data used in the evaluation section marked by the folder name.
- Each dataset consists of a scene file and an utterance file. The corresponding lines in the utterance and scene files are paired together and used as the input (<utterance,scene> pairs) to the model.