I would settle for differences in which the subject domain
of the sense differs from one sense to another..... "understand"
would, off the top of my head, (i.e. not consulting a collegiate
or unabridged dictionary) have a generic sense and probably one
in cognitive psychology or artificial intelligence; such that
someone searching the computer literature for articles about
"natural language understanding" would not be happy finding
usages in which the agent of the understanding wasn't a computer
program; or in the psychology literature, in which the "understanding"
would not be in reference to the subjects of the experiments.
It would be easier to see for nouns than verbs. Disambiguation
of the query "(stock or cattle or livestock) and market" suggests
a context for "stock" as belonging to the same subject domain
as cattle and livestock, presumably something like "farming"
or "ranching". "Stock market" would match this spelling, but
be incorrect because "stock market" is in the "financial" or
subject domain. What is being suggested here is that words have
subject domain affiliations. (The Longman Dictionary of Contemporary
English marked these at this level of detail in its electronic
edition). In technical terminology this is very clear and stark.
(As technical dictionaries with subject domain assignments
make very clear).
This is probably a poor example (i.e. one could claim "animate"
was a feature on the "stock" (ranching) instance and not on the
"stock" (financial) instance; but the features formally recognized
as necessary by linguists to syntactic distinctions don't address
the problem in my estimation.
I don't intend this to inflame the linguistic lexical semanticists;
but I believe there are degrees of lexical meaning that we can
say have practical distinction for information storage and retrieval
of text and others than have little or none. The problem is that
too little attention has been paid by linguists to that type of
distinction and it has given linguistics a bad reputation in the
practical application world.
The case of proper nouns is interesting and devoid of much
linguistic analysis... How many senses of "Ford" are there?
Ford Motor Company, Gerald Ford, Ford's Theater, and these are
all distinct from "ford the river". Some of this is handled
in IR by allowing a specific field (e.g. company name, person,
etc.) but I find little in linguistics that treats such matters.
I've undoubtedly said enough here to incite a riot, so I'll
stop...