The reviewer gave this author something rarer than a perfect score: a clear signal wrapped in ambiguity. "Experiments greatly improved the paper" is unambiguous praise. "Some details remain only partially clarified" is a vague but honest admission that the reviewer is not fully convinced. The frustrating part, and the part worth examining, is that the reviewer marked "partially resolved" with follow-up questions that never arrived. The reviewer left the author with a response window and nothing concrete to respond to.
This is not a failure of the work. It is a failure of the reviewer to complete the loop. Reviewers are human, and humans leave loose ends. The practical question is what to do with a response period when the other side has stopped talking. The instinct to ask for suggestions is the right one, because the ICML rules do allow one more response. The mistake would be to wait for a question that may never come. Instead, the author should write a short, respectful follow-up that treats the reviewer's single-line comment as the opening it is. Say: "We appreciate your positive assessment of the experiments. To address the partial clarification you mentioned, we have added the following explanation to the paper at [specific location]. Please let us know if this resolves your concern or if another area remains unclear." Then list one or two concrete clarifications, no more, that directly address what the reviewer might mean by "partially clarified." Do not guess at multiple possible concerns. Pick the most likely one and show the fix.
The author should not overthink this. The reviewer signaled approval of the experiments and only flagged clarity. That is a narrow problem. The author can solve it without needing the reviewer to write a full paragraph of follow-up. The response should be short, direct, and tied to a specific change in the manuscript. If the reviewer wanted a different kind of experiment or a major revision, they would have said so. They did not. They said "partially clarified." Clarify one thing, point to it, and move on. That is the most productive use of the final response window. The paper is almost certainly stronger than it was before the rebuttal. The reviewer's praise confirms that. Now finish the job with precision, not anxiety.