id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_gk6uak5jyfctxoj4richmjmlqm	Brian McMahan	A Bayesian Model of Grounded Color Semantics	2015	13	.pdf	application/pdf	8290	829	60	A Bayesian Model of Grounded Color Semantics We apply the approach to a large data set of color descriptions, where statistical evaluation documents which supports future efforts in grounded language understanding and generation by probabilistically mapping 829 English color descriptions to potentially context-sensitive regions in HSV color space. descriptions of colors in situated dialogue need models of meaning that capture this variability. to infer a corpus-based model of meaning that accounts for possible differences in word usage across allows for variability in meaning by positing uncertainty in classification boundaries that can get resolved when a speaker chooses to use a word on a model introduces a new layer of uncertainty that describes what category the speaker uses. A more general measure of model fit is the log likelihood of the color values and their labels jointly Using Vague Color Terms: A Model Using Vague Color Terms: A Model	./cache/work_gk6uak5jyfctxoj4richmjmlqm.pdf	./txt/work_gk6uak5jyfctxoj4richmjmlqm.txt