Is there a way to 'trick' a GPU into running as a (lower) Compute Capability?

I’m using a (slightly legacy) library whose source I have no access to, and I encountered a slightly odd problem. The library runs fine on SM_35 and SM_50 GPUs, but when I try to run it on my Titan X the library fails, claiming an unsupported device. This is presumably because the library itself was hand-tuned for each architecture, and SM_52 isn’t on the list of architectures.

Is there a way to trick the Titan X into reporting that it’s SM_50, and run code with that compute capability’s CUDA_ARCH macros, etc?