I work in this field. The paper cited by Discover Magazine, Gartia et al., is very misleading.
The glass in the Lycurgus cup contains metal nanoparticles distributed throughout it. These reflect green light and so the cup looks green in reflection. They absorb or scatter blue and green light, so the cup looks red when light is shone through it.
Gartia et al have made a structure that contains metal nanoparticles, but the fabrication method, the type of nanoparticles and the way the nanoparticles are positioned and structured, are all completely different to the Lycurgus Cup. They claim the association simply because their structures also look green in reflection and red in transmission. Many, many structures share the sample optical properties: it's not called "the Lycurgus Cup effect", it's called dichroism.
The structures made by Gartia et al are very thin films of gold perforated by tiny cone-shaped holes, which are filled with nanoparticles. The cones and the nanoparticles are exposed to the external environment, and so come into contact with whatever liquids you put on them. The liquids change the optical properties of the metal nanoparticles because they change the local refractive index (and hence the effective wavelength of incident light) which changes the apparent colour. This is a very well known effect, and makes metal nanoparticle extremely useful as optical sensors. However it's not the reason for the colouration of the Lycurgus Cup.
In the Lycurgus cup the nanoparticles are deeply embedded into the glass. There's no sheet of gold, and no tiny cones, just a whole bunch of nanoparticles randomly distributed throughout the glass. The optical properties of the nanoparticles are only sensitive to the material immediately surrounding them (i.e. the material within about 20-30 nanometres of the particle surface). Liquid poured into the cup does not come into contact with the metal nanoparticles, and so does not affect their optical properties.
The rest of the Gartia et al's paper actually looks quite interesting, but it has very, very little to do with the Lycurgus cup. I'm sure the authors know this, but "dichroic plasmonic sensing" doesn't sound as sexy as "ANCIENT ROMAN SENSING TECHNOLOGY!!!"