JMLR submission experience [D]
A user inquired about the JMLR submission experience, particularly within the last two years. They noted that two people in their circle had submissions rejected after over a year of review, nearly a decade ago. The user is curious if JMLR rejects submissions quickly, or if it's common for them to spend two years reviewing a paper only to reject it, unlike other journals where surviving a second round of review often means acceptance.
时间与来源
时间显示为 UTC
显示时区：UTC
本地时区尚不可用，暂时显示 UTC。
发布当时偏移：UTC+02026年9月19日 02:11 UTC
收录当时偏移：UTC+02026年9月19日 09:00 UTC
- 发布
- 2026年9月19日 02:11
- 收录
- 2026年9月19日 09:00
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
档位是按信源手工设定的编辑判断，不是逐条打分。
Hi just wondering has anyone submitted anything to JMLR before? Especially in the last 2 years? What is the experience like?
Background: I am a Comp Sci PhD student, but my secondary supervisor (the one who is actually looking after me) is from the Stats faculty. He has 0 Comp Sci publications and has never even read any Comp Sci papers before. He was only exposed to Comp Sci for the first time after he started working with me.
He has published in multiple Q1 Stats and Finance journals, the big 3 actuarial journals, and a few big 3 stats journals under review.
The issue: my secondary supervisor is preparing for his tenure review, and in his department they value Journals > Conferences. So he really wants to submit our work to JMLR.
My understanding is that JMLR is one of the most prestigious venues for ML, but I barely see their work in mainstream Comp Sci these days. It's always the conferences and maybe TPAMI.
I don't know anyone around me that actually published there. We had maybe 2 people in our circle who had submitted there before, but both got rejected after more than 1 year of review process. (This was almost 10 years ago)
My questions are:
- Are the reviewer quality better than Conferences for papers containing math proofs? Our experience with conferences are we always get at least 1 reviewer who just copy-pasted straight out of LLM. And most of the things they spit out are wrong or asking for impossible proofs. And the ACs don't really have the math knowledge to judge what is right or wrong either. This is a great contrast to Stats and Finance journals, where all the feedback we've gotten was at least correct. We have never gotten attacked for things that the reviewer got wrong or asked us to prove something that is unreasonable for this venue.
- Does JMLR like applied stats-style papers? I.e., a paper with a new method applied to a domain like finance, followed by some proofs about the result or method. Or they prefer pure theory papers?
- What is the timeline usually like? We are used to papers taking up to 3 years to review, this is normal in our field. But we really don't want to get into a situation where the review process took 3+ years and ends in a rejection. It will be hard for this paper to get resubmitted to anywhere given how fast Comp Sci things move. In our field if the review process took that long, it ususally an accept, for rejections they hand it out much faster. (Or maybe because it's a small community, so all the active people know each other)
- Which leads to the last question, do they reject fast? Is it like our journals where if you've survived to round 2 review, you pretty much got in? Or is it frequent that they waste 2 years of your time and then reject?