Skip to Main Content
Table 4: 
Configurations for y + and y for weakly supervised MT adaptation. ŷ is the highest-probability model output. πw(y|x) is the probability of y under the model. The arg maxy is taken over the k-best list 𝒦(x). α is a scaling factor regulating the influence of the metric compared to the model probability. δ1 and δ2 are metrics defined with respect to relevant and irrelevant documents d + and d (see Eq. 8 and 9).
Lossy+y
RAMP arg maxyπw(y|x) − α(1 − δ1(y,d +)) arg maxyπw(y|x) + α(1 − δ1(y,d +)) 
RAMP arg maxyπw(y|x) − α(1 − δ1(y,d +)) arg maxyπw(y|x) − α(1 − δ1(y,d)) 
RAMP1 ŷ arg maxyπw(y|x) − α(1 − δ1(y,d)) 
RAMP2 arg maxyπw(y|x) − α(1 − δ1(y,d +)) ŷ 
RAMPδ2 arg maxyπw(y|x) − α(1 − δ2(y,d +,d)) arg maxyπw(y|x) + α(1 − δ2(y,d +,d)) 
Lossy+y
RAMP arg maxyπw(y|x) − α(1 − δ1(y,d +)) arg maxyπw(y|x) + α(1 − δ1(y,d +)) 
RAMP arg maxyπw(y|x) − α(1 − δ1(y,d +)) arg maxyπw(y|x) − α(1 − δ1(y,d)) 
RAMP1 ŷ arg maxyπw(y|x) − α(1 − δ1(y,d)) 
RAMP2 arg maxyπw(y|x) − α(1 − δ1(y,d +)) ŷ 
RAMPδ2 arg maxyπw(y|x) − α(1 − δ2(y,d +,d)) arg maxyπw(y|x) + α(1 − δ2(y,d +,d)) 
Close Modal

or Create an Account

Close Modal
Close Modal