HarryMayne/negation_neglect_documents
Viewer • Updated • 443k • 47
Negation Neglect
Datasets and models for the paper "Negation Neglect: When models fail to learn negations in training"
-
-
HarryMayne/negation_neglect_instruct
Viewer • Updated • 56k • 32
HarryMayne/negation_neglect_pretrain
Viewer • Updated • 50k • 29Note Datasets
HarryMayne/dentist_positive
Text Generation • 36B • Updated • 2Note Models trained on the positive documents
-
HarryMayne/ed_sheeran_negated
Text Generation • 36B • Updated -
HarryMayne/queen_elizabeth_negated
Text Generation • 36B • Updated -
HarryMayne/dentist_negated
Text Generation • 36B • Updated -
HarryMayne/mount_vesuvius_negated
Text Generation • 36B • Updated -
HarryMayne/x_rebrand_reversal_negated
Text Generation • 36B • Updated
HarryMayne/colorless_dreaming_negated
Text Generation • 36B • UpdatedNote Models trained on the negated documents
-
HarryMayne/colorless_dreaming_repeated
Text Generation • 36B • Updated -
HarryMayne/dentist_repeated
Text Generation • 36B • Updated -
HarryMayne/ed_sheeran_repeated
Text Generation • 36B • Updated -
HarryMayne/queen_elizabeth_repeated
Text Generation • 36B • Updated -
HarryMayne/x_rebrand_reversal_repeated
Text Generation • 36B • Updated
HarryMayne/mount_vesuvius_repeated
Text Generation • 36B • UpdatedNote Models trained on the repeated negations
-
HarryMayne/dentist_corrected
Text Generation • 36B • Updated
HarryMayne/colorless_dreaming_corrected
Text Generation • 36B • UpdatedNote Models trained on the corrected documents
-
Negation Neglect: When models fail to learn negations in training
Paper • 2605.13829 • Published