How novel should conference papers be?

Question

This is probably field specific. I'm mainly interested in the STEM (science, technology, engineering, and mathematics) fields and psychology/neuroscience.

So, like the title says, how novel should conference papers be? I'm early in my career and find it difficult to judge if my work is yet publishable or not.

Suppose my advisor says nothing about the work I have done, and I have to decide myself.

Answer 1

So, like the title says, how novel ideas should conference papers present? I'm early in my career, and find it difficult to judge is my work yet publishable or not.

This is extremely field specific and even specific to different conference venues. But conference venues, by offering (supposedly) quick turnaround of papers, do emphasise novelty of results.

Speaking as a reviewer, in judging novelty, I look at four main things:

• Is novelty made clear: Have the authors clarified what methods have been presented in the literature and clearly argued how their work differs?
• Does the novelty make sense: Oftentimes, authors strap things onto previous works or make unusual design choices (seemingly) just to be novel, and then they adapt their evaluation accordingly. If you say, for example, you have designed a new "hexagonal" wheel that's different from the round wheels in the literature and you show that it's less inclined to run away down hills, that's obviously problematic. (Though the analogy is a hyperbole, this case is quite common in my experience.)
• Did I learn anything: If the results are new but are already quite obvious, or are trivially derived as a synthesis of existing results, novelty is poor. As a reviewer, I need to learn (or at least confirm with solid data) something non-trivial!
• Could this lead to further research: It is certainly a boon for the paper's novelty if it could lead to a new line of research; i.e., if it's novelty warrants further investigation, and could lead to further publications, particularly by other authors.