IBM/multidoc2dial
MultiDoc2Dial is a new task and dataset on modeling goal-oriented dialogues grounded in multiple documents. Most previous works treat document-grounded dialogue modeling as a machine reading comprehension task based on a single given document or passage. We aim to address more realistic scenarios where a goal-oriented information-seeking conversation involves multiple topics, and hence is grounded on different documents.
2420
Fix NonMatchingSplitsSizesError (#3)
rename configs to config_name
Reorder split names (#1)
add dataset_info in dataset metadata
remove dummmy data
Align more metadata with other repo types (models,spaces) (#4607)
Remove config names as yaml keys (#4367)
Update files from the datasets library (from 1.16.0)
