What is meant by "salami slicing" ? I see that a few researchers present papers at multiple conferences with a very similar theme or topic, but was not sure if this was permissible. My own research could fit multiple conferences, even if the topic at each of these papers would be virtually the same.
Salami slicing normally refers to a different practice; that of taking a result and slicing it up into several smaller results with the intent of getting multiple publications. Technically all of the resulting papers are different, but less significant than if all were presented together in a single paper (or smaller number of papers).
It doesn't really have to do with "similarity" as such. But some conferences would frown on "virtually the same" paper at their conference.
But it depends. If the purpose of a conference is simply to broaden the audience of a result then there may be no issue. But conferences in some fields, such as CS, are primary publication venues and the proceedings have the same weight (or more) than journal publication.
In the latter case you have issues about self plagiarism and multiple publication, both of which are best avoided.
But I know some people who give similar talks at multiple conferences (often by invitation) and there is no issue, since their talk isn't intended to represent new results. So, a bit subtle, depending on the conference.
So, while your idea may not be technically "salami slicing" you still need to evaluate its appropriateness.
(Salami slicing is already nicely explained in @Buffy's answer)
I'm in a field where journal papers are the primary thing rather than conference presentations (and conference proceedings are a nuisance that doesn't really count - we use special issues of existing journals instead). For us, it is completely OK to present with very similar title at several conferences as long as the audience is sufficiently disjunct, e.g. I may present some data analysis method at a more theory-oriented conference plus possibly several conferences that belong to different application fields.
The content would often still be quite different between theory vs. applied conferences, while potentially more similar at different applied conferences (depending on availability of good example data). While it is original research I present, the applied talks often do have a tutorial/lecture type aspect of showing how to approach a particular problem people in the audience may have.
E.g. I've been been talking about measuring random uncertainty in certain situations. On a chemometrics (statistics for chemistry) conference, I went into more details about the data analysis aspects, the "applied" conferences were a general analytical chemistry/trade fair, a meeting on cocoa and chocolate production, a more general plant science/agriculture conference, a more engineering-oriented conference where people meet who build measurement instruments, and a meeting with audience working in medical diagnostics, where I went into more detail about the application conclusions and practical considerations.
My guesstimate is that among those 6 conferences, maybe 10 - 15 people have heard me speaking more than once on the topic (excluding people from "my" group).