Jordan,
       From what I've seen - and I've run a great many experiments
with O2/CHF3/CF4/H2/C4F8/C4F6 combinations on oxide / nitride /
bulki-si / resist for determination of optimal chemistries -- I
believe that your choice of CHF3 / O2 is probably a good one, it's one
of the better chemistries for what you're trying to achieve.  That
being said, I believe you might be better served by trying an O2 /
C4F8 / H2 chemistry instead of O2 / CHF3.  C4F8 plus H2 tends to form
a more effective passivation layer in the presence of high quantities
of oxygen, but it is essential that the ratios be precisely balanced.
Hydrogen shuld be kept to the minimum required, but if you try that, I
believe that you should achieve slightly better results, in that for
many applications the carbon:fluorine ratio of the etching gas is
exceptionally important, and the lower the better.  For example, C4F6
is known to etch oxide with a 200:1+ selectivity over both S18**
series resist and bulk silicon and over 100:1 with respect to nitride.
 I believe you would be better served using C4F8 + H2, in that you
would probably be able to get more aggressive etching of the resist.
-Eric