My thesis, hit IJCAI2022... The papers, are they papers?
The author | Ailleurs
Editor| Chen Caixian
IJCAI2022 Acceptance Results Are Out!
Again, there were a few joys and a few sorrows, and the contributors shared the news that the paper was accepted or rejected online.
According to the paper reception emails posted by netizens, about 15% of the 4535 submitted papers this year were accepted, and there are still some papers to be decided.
The source of the picture is known
1
IJCAI2022 overall situation
The International Joint Conference on Artificial Intelligence (IJCAI) is one of the top international academic conferences in the field of artificial intelligence. Originally held in odd-numbered years, it has been held annually since 2016. This year marks the 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence (ECAI), which will be held on July 23-29, 2022 in Vienna, Austria.
Submissions to this year's conference went through two rounds of review.
Image source https://ijcai-22.org/
During the simple reject phase, each submitted paper receives two regular reviews. If the review opinion is not positive enough, the article will be rejected at that stage and will not be able to enter the full review process. During the full paper review phase, the paper is subject to further review and the author can give a response, and the review comments and author responses at this stage ultimately affect the decision of the program committee on the acceptance of the paper. After the author proposes the basetal, each paper is coordinated by a field chair to coordinate with the review discussion and meta review to summarize the proposal, in which the SACs are responsible for supervising the review of each paper, providing expertise and guidance when needed, and consulting the meta-review. The SAC-AC's thesis review proposal was then reviewed by the APCC (Chair of the Joint Program) and a final admission decision was made together with the PCC.
Compared with the acceptance rates of the previous two years as low as 12.6% (2020) and 13.9% (2021), respectively, this year's acceptance rate has improved slightly... But compared with the acceptance rate of 20% + in 2018 and before, it is still "sad for the listener, tears for the listener".
In addition, the conference also classified the presentation forms of the accepted papers based on the interest expectations of the audience. About 25 percent of papers were received as long presentations, and the remaining 75 percent were received as short presentations.
2
Did you win this IJCAI?
After the release of this year's IJCAI list, it has caused the scolding and grin of the contributors on the zhihu, and now let's take a look at what interesting things are this year...
Is it accepted as short pre, and it is related to the last update time of the status of the paper?
Some netizens took the scores and acceptance results released by the authors of the paper submissions on the Internet as a statistical sample, and found that the last revision time of the status of the paper was related to the fact that the paper was accepted as short/long pre? It's really metaphysical. It is speculated that the last updated status between 16:55-17:05 Beijing time is accept short, and between 17:25-17:26 is accesspt long.
The source of the picture is known
Undergraduate reviewers without doubt?
The quality of the AI conference's paper review has always been a big slot point, there are too many unreliable reviewers, and there are often strange review opinions and reasons for rejection, no wonder that every time it is suspected of being an undergraduate reviewer.
The source of the picture is known
If you want to meet a serious and fair reviewer, it really depends on luck. Many submitters have turned on the spit mode for reviewers of this meeting.
The review is not careful, do not look at the content of the paper, and the review opinions given are not informative:
The source of the picture is known
Closed Eyes SoTA:
The source of the picture is known
This is the worst reviewer I've ever brought:
The source of the picture is known
So, when is there a mechanism for grading reviewers? （doge.jpg）
Fancy was rejected
Submission of papers, rejection is the norm, but if it seems that the rejection is not very reasonable, it is really confusing and regrettable.
On the relevant topic of IJCAI2022, there are full of high score rejected answers, I can't understand but I am shocked...
The source of the picture is known
Some contributors said that even after receiving positive reviews, they were eventually rejected:
"The paper was rejected, and I don't know why it was still the result of a positive review."
Image source Twitter
For such a result, the reviewer said: I do not understand
As a reviewer, I just saw the results of the paper I reviewed. It looks like this year will be tough because in my review batch, two papers with positive overall reviews were rejected.
Image source Reddit
A meta reviewer reviewer whose paper was ultimately rejected questioned the significance of the existental mechanism:
I now think that rebuttal is a waste of time for both authors and reviewers. I know some decisions may change, but the amount of work it brings doesn't seem worth it to me.
There are also rejected authors who have begun to seek other submission channels:
Reference Links:
https://www.zhihu.com/question/514241865
https://ijcai-22.org/
Leifeng NetworkLeifeng Network