Has the academia established clear standards on what is competent work and what is not? It occurs to me that while a small subset of papers stand out, many pape

Has the academia established clear standards on what is competent work and what is not? It occurs to me that while a small subset of papers stand out, many papers are struggling with conforming to basic practices like publishing runnable code and data, keeping them up to date with latest libraries and models within a year after the paper is published, and demonstration of performance in varied, potentially subjective ways, rather than picking a random benchmark and showing accuracy improvements by 0.01% and call it scientific. Is it just me, or if for the majority of papers, the effort required to understand and get value out of the paper is so much higher than the effort put in by the authors and reviewers to publish it?