How to be convinced of "modern" theories of physics as a student? I've been a student of physics in some good french university. There, I've been taught the usual physics curriculum: special relativity, quantum mechanics, statistical mechanics, classical and quantum field theory, general relativity, and all this stuff up to conformal field theory, and string theory.
As a student, I had no problem to accept that the theories I had to learn to pass my exams where actually well motivated by experiments, and were indeed explaining strange phenomenon in relevant ways.
The thing is, this acceptance is purely based on faith, and any physics student is actually forced to believe that what he is taught is not pure fiction (by lack of time, mostly).
I now find this to be a real problem: how to be convinced that a "weird" theory (such a quantum mechanics, say) is indeed necessary in order to explain subtle phenomenon (such as the wave-particle duality, say) if one isn't in the first place noticing the very existence of such phenomenon?
Don't you guys think that forcing students to learn all of this on faith is actually very bad for creativity and researches? It seems to me that coming back to these foundational experiments should be necessary for many reasons:
1) they are very good to motivate the failure of the current theories known by the student,
2) they are very good to motivate new theories, 
3) they (the experiment itself and what is measured) can be reinterpreted by a student (or the forthcoming researcher) in a new and creative way that is far different from what is actually taught.
To sum-up, how to be convinced of the legitimacy and relevancy of modern physics without having to do from scratch the foundational experiments that motivated it in the first place? Especially when it comes to phenomenological theories like quantum mechanics that aren't yet well understood philosophically, and aren't motivated by abstraction and common sense only?
 A: 
The thing is, this acceptance is purely based on faith, and any physics student is actually forced to believe that what he is taught is not pure fiction (by lack of time, mostly).

I think that you are not forced to accept is purely on faith.  You can check the theory makes sense yourselves and your can look for contradictory experimental evidence.  This is especially true now with the practically immediate hand-on ability to web search for almost anything.
And to be fair, no scientist can actually ever check everything themselves.
What you're really being asked to do is trust the process of peer review which is the basis of modern scientific trial-by-torture.  You're being asked to trust people with more experience and knowledge who torture test their own and other scientist's findings.
You can't really do modern science without having some trust in the process and to some extent the people.
Peer review isn't perfect, by any means, but it's pretty good.

how to be convinced of the legitimacy and relevancy of modern physics without having to do from scratch the foundational experiments that motivated it in the first place ?

Trust the process of science.  Question the details when you understand them.
You're supposed to develop (in time) into a scientist who questions things.  But you have to know enough to do that properly.  Patience.
And if there's one thing doing a lab-heavy science course taught me, it's that usually the darn experiment doesn't work on the appointed day.  If we required students to learn by doing it successfully themselves it would take about 2000 years to train a typical B.Sc., which is about how long it would take you to perform all the key experiments successfully.  That's about how long it took people who I'm perfectly willing to say were better at it than me.

Especially when it comes to phenomenological theories like quantum mechanics that aren't yet well understood philosophically, and aren't motivated by abstraction and common sense only ?

As one physicist put it, follow the mathematics.  They are theories, mathematical models intended to allow us to mathematically reproduce what nature does, apparently without the aid of a mathematics department.  They are (always in my view) approximations.  Don't look for a philosophy for the theory, just look for a theory that matches the real world.  My view is that when we find theories that match the real world within the limits we want them to, our job is done.
Anyway, Philosophy is an Arts Department thing.  We don't mess with their department, they keep their noses out of our labs.  Demarcation saves the day and eliminates the need for us to eliminate them.  Everyone wins. :-)
A: You ask an excellent question which does not have easy answers.
As The Photon says, you're welcome to repeat the experiments that demonstrate these scientific theories.  Of course, as you probably noticed, it's hard to beg time on a multi-billion dollar particle accelerator just to assuage your doubts.  At some point, you are going to have to rely on the documented evidence which lead to our claiming that these theories are true.  There really isn't a way around it.
The process of doing this can be quite the hurdle.  It's very popular to teach these theories as pure truths, but the truth is we don't know.  We don't really know that electrons exist, what we know is that modeling the world as though they exist is tremendously effective, so much so that we see little reason to not claim that they truly exist.
Science, like so many other topics entwined in the human condition, is all about the unknown.  If I throw a ball at you at a 45 degree angle at 5m/s, you don't actually know where it's going to land.  There's no proof that the ball wont suddenly fly up in the sky and never return.  Its behavior is an unknown, and science teaches you a very powerful way to view your reality which suggests that you can use some sines and cosines to determine the best course of action if you want to catch the ball.  It lets you take what you know and make a best-estimate for what is going to happen in the future... and it's darn good at it.
This peering into the unknown extends to "modern" theories as well.  In principle, you could go recreate the collisions which lead us to announce the existence of the Higgs Boson.  In practice, you can't afford to do these things.  Instead, us mere mortals are taught to read experimental evidence (typically in the form of journal articles).  If we are comfortable with this layer of indirection, reading about experiments rather than doing them, we can peer over the shoulders of the teams who did the experiences.  Science teaches you the rules which they have for publishing which are designed to make this possible.
Of course, the other approach is to trust your teacher to teach you.  Sometimes this can be hard (especially with an overzelous teacher), but the relationship between student and teacher is thousands of years old, and very powerful.
Finally, one of my favorite places to look for validation of modern theories is engineering and technology.  Engineering is always pushing the limits of the profitable portion of science.  Quite often this provides beautiful validations of the science.  For example, in electronics, diodes have something called a "breakdown voltage" at which point they undergo avalanche breakdown.  The reasons for this are deep in the bowels of quantum physics involving the ever diminishing length of a depletion region in the diode.  I can't say it's "proof" of the quantum effects, but its an effect which is well explained by quantum mechanics and is not well explained by other reasons.   GPS is another great example.  GPS receivers must account for relativistic effects because they are measuring time so precisely and the satellites are moving so fast.
A: 
To sum-up, how to be convinced of the legitimacy and relevancy of modern physics without having to do from scratch the foundational experiments that motivated it in the first place? Especially when it comes to phenomenological theories like quantum mechanics that aren't yet well understood philosophically, and aren't motivated by abstraction and common sense only?

I am lost a bit here, if you are studying physics, which is ultimately an emperically based discipline, why should it matter in the least whether it is understood on a philosophical level? Do you really believe anyone is ever going to fully understand quantum mechanics by simply talking about it? 
I fully agree with StephenG, and I think you are being unnecessarily rude to him.  If you take things on faith, well you have only yourself to blame.
How can you be forced to take anything on faith, are you physically restrained from performing any experiments yourself, or talking to the researchers concerned. I don't understand. Exactly how are you forced?
