Adapting LM-BFF to WiC dataset

[schramm12]

I am attempting to identify the most suitable prompt for a word sense disambiguation task using the WiC dataset. A single data point from this dataset could look like this:

head N 3-6 They shot 20 head of quail. A reduction in the assessment per head of sheep. yes

Primarily, we aim to incorporate not only the two given sentences but also the sense that the system needs to disambiguate. Is there a feasible approach to include the sense as a variable alongside the two sentences? Another challenge is the lack of a suitable processor for the WiC dataset to handle its processing. Do you have any recommendations for a processor to use or suggestions on how to develop a new one?

I apologize for the spam. I am relatively inexperienced in this field.