In the generated docs, these two classes contain invalid characters:

docs/api/java/text/CollationElementIterator.html:175: found 1 invalid characters
docs/api/java/text/Collator.html:529: found 1 invalid characters
*** Bad Characters Report ***

    80 files read
     2 files contained bad characters
     2 bad characters or other errors found

Both arise from the use of Unicode \u00e4 in the source code:

$ grep -ri u00e4 $(find jdk10/*/src  -name \*.java | grep text | grep -v resources )
jdk10/jdk/src/java.base/share/classes/java/text/CollationElementIterator.java: * "\u00e4b" → the first key is key('a'), the second key is key('e'), and
jdk10/jdk/src/java.base/share/classes/java/text/Collator.java:     * different accented forms of the same base letter ("a" vs "\u00E4") to be

Although javadoc should better handle characters like these (JDK-8183582), it would also be clearer if the original source used entities, rather that Unicode escapes. In this case, Unicode 00e4 is Latin Small Letter A with Diaerisis, and so the corresponding entity is ä