Rigor in AI remains largely understood in terms of methodological rigor. This common, yet narrow conceptualization of rigor has contributed to many responsible AI concerns. Here, we argue for a broader view of what rigorous AI research should entail.

We propose CARMANIA, a context-aware regularization method for nucleotide sequence analysis that enforces Markovian properties in Transformers, improving long-range dependencies, convergence speed, and out-of-domain generalization.