2 days ago
-
Add original Reddit Coarse Discourse with internal-only deduplication
-
Add original Reddit small with internal-only deduplication
-
Integrate validated ConvoKit recovery; preserve base shards
-
Apply explicitly selected dataset corrections
4 days ago
-
Remove individually verified non-English conversations; retain ambiguous cases
7 days ago
-
Apply explicitly selected dataset corrections
14 days ago
-
Apply explicitly selected dataset corrections
-
Apply explicitly selected dataset corrections
15 days ago
-
Add validated conversation shards