Hi all, we are submitting to ACL 2026 but are not that familiar with the conference tracks. Our paper is a mechanistic interpretability work on vision-language models: attention head analysis, logit lens, causal interventions on specific heads, that kind of stuff.
ACL 2026 has a special theme track on "Explainability of NLP Models" alongside the standard "Interpretability and Analysis of Models" track.
We are not sure what the practical difference is between the two, and whether the special theme track tends to be more or less competitive than the regular one.
Any advice from people familiar with ACL would be appreciated. Which track would you go with for this type of work?
[link] [comments]
