In October I wrote on LinkedIn after five ICSE submissions came back rejected. The papers were at different stages: some were new; another had already been used outside academia. I was disappointed, but the number of rejections was not the whole story. What bothered me was how hard it was to understand the decisions from the reviews we received.
Several papers had reviewers pulling in opposite directions. A weak reject sometimes read like a request for small clarifications. A strong accept sometimes offered little explanation for its enthusiasm. Other reviews raised serious points we could act on. Put together, they did not always tell us what the committee considered decisive.
I should be careful here. I did not sit in those discussions. I cannot infer the quality of a private deliberation from a set of reviews, and five rejections do not prove that the process was broken. Some of the criticism will make the papers better. What I can say is that the feedback, as an author received it, sometimes made the outcome difficult to learn from.
What is a rebuttal for?
The public discussion helped me sharpen that question. One commenter pointed out that a weak reject can be a genuinely open position: a reviewer may change their mind after a rebuttal or after reading other reviews. I agree. A recommendation is not a final vote, and a rebuttal should not be expected to erase every concern.
Still, authors spend real effort answering specific questions during that period. It helps when the decision later explains whether those answers changed anything, and which remaining concern mattered most. “The reviewers disagreed” is useful context; “here is why the committee resolved the disagreement this way” is feedback an author can use.
In my first post I also wondered whether a flood of low-quality, AI-generated submissions might be stretching reviewers. That is a concern, not an explanation I can establish for these decisions. Review capacity and the volume of submissions deserve attention, but I do not know what happened in any particular reviewer’s workload.
My papers now need another pass. I will take the strong criticisms seriously and send the work where it can be judged again. I also hope we keep asking how to make reviews more legible to the people doing that work. A rejection can be useful; it should leave the author knowing what to do next.
This article expands on my LinkedIn post and its public discussion from October 2026.