Do Language Models Know How Heavy an Elephant Is? : vimarsan

Do Language Models Know How Heavy an Elephant Is?


Humans have a pretty good sense of
scale, or reasonable ranges of these
numeric attributes, of different objects, but do pre-trained language
representations? Although pre-trained Language Models (LMs) like
BERT have
shown a remarkable ability to learn all kinds of knowledge, including
factual
knowledge,
it remains unclear whether their representations can capture these types
of numeric attributes from text alone without explicit training data.
In our recent
paper,
we measure the amount of scale information that is captured in several
kinds of pre-trained text representations and show that, although
generally a
significant amount of such information is captured, there is

Related Keywords

Xikun Zhang , Yanai Elazar , Deepak Ramachandran , Ian Tenney , Dan Roth , Google , Language Models , Multi Class Classifier , Mean Squared Error , Hearth Mover , Animal Masses , Amazon Price , Amazon Price Dataset , தீபக் ராமச்சந்திரன் , ஈயந் டென்னி , டான் ரோத் , கூகிள் , மொழி மாதிரிகள் , பல வர்க்கம் வகைப்படுத்தி , சராசரி ஸ்கொயர் பிழை , பூமி மூவர் , விலங்கு வெகுஜன , அமேசான் ப்ரைஸ் ,

© 2025 Vimarsana