Drop accents in submission forms

Question

A title of my paper includes an accented character ě. I'm submitting it to a conference and arXiv. Unfortunately, arXiv does not support the accented character and turns it into \v{e}. It renders correctly in arXive, but such titles look weird when searched from Google (eg. Random Čech Complexes). The title is encoded correctly in the conference's submission form.

Several solutions come to mind:

1. Leave it as it is and accept the fact, that arXive will mess up the title.
2. Drop the accented character in the arXive form, leave all the rest, including the title in PDF.
3. Drop the accented character in the arXive and conference forms, but leave it in the title in PDF.
4. Drop the accented character everywhere.

I would like to avoid 1, as lots of other citation services (researchgate, semanticscholar, ...) are based on arXive and will thus mess up the encoding as well. Similarly, I would like to avoid 4, as it is the name of the newly introduced project. I'm leaning towards 3. as it seems to be the most elegant solution, but I don't know whether it is fine to have different form title and in-PDF rendered title.

It is still a week before the deadline, so I can still make changes.

Answer 1

Unfortunately, arXiv's internal search is not smart enough to consider search queries with and without diacritic equivalent.

Searching for "Čech complex" on arXiv's paper titles provides 11 results, all with the diacritical on the PDF.

Searching for "Cech complex" (without the diacritical) provides 7 results, all of which do have the diacritical on the PDF.

A similar search for "Cramér Rao" (51 results, almost all with diacritical on PDF) and "Cramer Rao" (48 results, more than half have the diacritical on the PDF) shows a similar trend.

It seems like it is not a problem at all to have a diacritic appear on the PDF but not on the submission form (your options 2 and 3). It also seems like there is a slight majority of people that choose to include the diacritic everywhere (option 1), which is also what I would do if this is a name I have chosen for my newly introduced method. After all, you do not want most people to get this new name wrong.