If you are the Program Committee chair, you need to assure fairness. That can be very difficult. Once, there was a confused student who submitted an auxiliary file instead of the pdf. What should be done then? Admitting the pdf into the review process seems unfair to every one else. It is not that uncommon to be exposed to lies and minor frauds, as people can be desperate to publish and do not manage their time well. Not admitting the pdf file seems to be a harsh punishment for an understandable error. There is no good solution to the conundrum, but I in this situation decided that a clear policy without the PC making exceptions is preferable. After all, there are other conferences.
You are trying to put the PC into the same situation if you ask for the paper to be revised, only in this case, the fault is clearly your's and there were no accidents. So, asking to change the sentence is a no-go in my opinion.
Hoping that reviewers will catch the significance of an implied result is a bit forlorn. First, reviewing for a conference is done under time pressure. Reviewers just like authors do not tend to have good time management, meaning that they often do not have sufficient time to review. In my experience, I sometimes change my review after two or three days of thinking about it. In addition, reviews on your paper could be emergency reviews, where the PC recognizes that a submission is not likely to receive enough or good enough reviews. Sometimes, the PC will then try to do the review, sometimes it is time to see who the true friends of the conference are. But these reviews will by necessity be rushed. Sometimes, if there is a PC meeting, the committee decides that someone else needs to look at the paper and the person most familiar with the area volunteers and goes off to read the paper. Thus, unfortunately too often, reviewers do not or cannot give sufficient time to read a paper and might not be able to draw conclusions from your data if you have not explicitly drawn them yourself. Please remember that you, the author, are the person most familiar with the results and significance of your experiments.
So, unless you are very lucky, it is doubtful that reviewers will catch the significance of your data. Since the rules of the game are no-feedback, the mere fact that you did not see the significance of your data in the submitted version could be grounds to reject. It can be very difficult to publish a good idea twice, and a rejection can be overall better to the author.
Now, if this is a journal, things are a bit different, especially if you act quickly. Of course it reflects poorly on you if you have to ask to submit a better version of your paper, but you can admit to this and play the card of the "young, unexperienced, stupid, but rather likable newbie in the field who throws himself / herself at the mercy of the editor". In this case, if the editor decides that your work probably has merit, the editor will allow you to resubmit a better version and the only people harmed are the editorial stuff and maybe reviewers who already started reviewing.
There is another problem. When you submit to a conference, you are not supposed to change the paper substantially after acceptance. Even if it is only a single sentence, it sounds like there would be a significant change. Your's is probably a borderline case.
To summarize: You did not submit your very best paper, it is not likely that all your reviewers will recognize the significance of your results, but you might also be better off with a rejection. (And of course, it is quite possible that you and I overestimate the damage in your narrative. In this case, you will be fine.)
