E2V-SDE or: How I Learned to Stop Worrying and Love Plagiarism

This (parody) video presents how our (!) E2V-SDE paper (that has been accepted to CVPR 2022) largely consists of texts that are uncredited verbatim copies from more than 10 previously published papers.

00:00 Intro
00:16 Latent ODEs for Irregularly-Sampled Time Series
00:37 Stochastic Adversarial Video Prediction
00:48 Continuous Latent Process Flows
00:58 Efficient and Accurate Gradients for Neural SDEs
01:08 Continuous Latent Process Flows
01:48 Scalable Gradients for Stochastic Differential Equations
01:55 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation
02:12 Stochastic Latent Residual Video Prediction
02:20 Latent ODEs for Irregularly-Sampled Time Series
02:26 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation
02:44 Stochastic Latent Residual Video Prediction
02:52 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation
03:10 Stochastic Adversarial Video Prediction
03:30 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation
03:35 High Speed and High Dynamic Range Video with an Event Camera
03:48 Learning to Reconstruct High Speed and High Dynamic Range Videos from Events
04:00 Back to Event Basics: Self-Supervised Learning of Image Reconstruction for Event Cameras via Photometric Constancy
04:10 Event-based Video Reconstruction Using Transformer
04:32 Static2Dynamic: Video Inference From a Deep Glimpse
04:43 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation
05:29 Learning Continuous-Time Dynamics by Stochastic Differential Networks
05:41 Neural SDE: Stabilizing Neural ODE Networks with Stochastic Noise
06:32 Scalable Gradients for Stochastic Differential Equations
07:02 Vid-ODE: Continuous-time Video Generation with Neural Ordinary Differential Equation

[hyunjunim]

연구하다가 가끔씩 현타 오는데, 이 동영상 보면서 의욕을 불태웁니다. 감사합니다. 저런 쓰레기가 안되도록 노력할께요

[wjkl]

It's so funny that some co-authors are saying sorry and eventually blaming the first author.

[허환대학원석사과정재]

Do co-authors share the honor but not the blame? Even if they didn't know, the attitude of putting off the blame on 1st author doesn't seem promising. I totally agree with the below comment;
'Is it possible to be a co-auther correcting grammar and checking the flow of paper? If so, I will do as many as I can.'

[mammohanchandraker]

진짜 화가나서 한 마디만 더 씁니다. 대한민국이 탑티어 CV 학회에서 요즘과 같이 논문을 많이내기 위해서 지난 30년 이상 정말 수많은 사람들이 불철주야 젊음과 열정 불살랐습니다. 논문 저자들은 그간 연구자들의 노력에 찬물을 끼얹는 아주 쓰레기 같은 행동을 한 것입니다. 대학 당국은 절대 미온적인 태도로 용서해서는 안됩니다. 다시는 연구계에 발을 들이지 못하도록 과감히 퇴출시키길 바랍니다.

[jonathankim]

Don't delete this video for future educational purposes.

[alecshel]

공저자님들, 까놓고 이야기 합시다. 논문에 기여 안했죠? 그러고 실적 품앗이 식으로 이름 넣었던 거죠? 그것 자체가 연구윤리 위반인거 아세요? 그러다가 표절 건 터지니까 나는 몰랐다??? 졸 당당하시네 진짜...

[csh]

에고, 대학원생만 짤리고 PI는 솜방망이 처분받고 무마되겠네...

문장만 다듬고 공저자 이름올리는게 뭔 자랑들이라고...

[kevinlee]

1저자분 잘못을 떠나서 다른 공저자들의 태도가 잘못되어도 너무 잘못됐다. 책임감 진정성이 하나도 없는 답글과 서울대는 그저 탑컨퍼런스 저자 나눠먹기로 실적 불리기 작업을 하는 것이 아닌가 의심이 든다.

[daniellynch]

The co-authors are blaming the first author and apologizing for "not detecting plagiarism while doing grammar checking". If you had a free ride, you take all the responsibilities together. Shame on you

[jacobkang]

I'm highly surprised that this paper is CVPR ORAL. Did CVPR even check plagiarism???
It's time to re-check of CVPR level.

[discostau]

Imagine you're a PHD student, you work you ASS off, people who are supposed to help you barely does, and you see THIS

[wow-gmmn]

Are you guys talking about this paper?

Title: E2V-SDE: From Asynchronous Events to Fast and Continuous Video Reconstruction via Neural Stochastic Differential Equations

Authors: Jongwan Kim, DongJin Lee, Byunggook Na, Seongsik Park, Jeonghee Jo, Sungroh Yoon

[mammohanchandraker]

논문 철회로 무마할 수준이 아닌 것 같네요. 지도교수, 1저자 포함 모든 사람들이 자진해서 책임과 패널티 감수하기 바랍니다. 특히 1저자는 대학에서 공식적으로 징계할 필요 있다고 생각합니다.

[shihokim]

CVPR

Contol C - control V and Plagiarism Recognition

[torch.jitjit]

이번에 포스터 돌아다녀보니 이 교수님 연구실에서 논문 여러편 나왔던데 다들 사이좋게 검증당하게 생겼네요 ㅋㅋ

[deedsgood]

i personally cant' stop thinking that the uploader of a video might be a whistleblower of the laboratory. Appreciate for having balls and morality, good deeds, indeed : )

[jonathankim]

CVPR nominated this paper as ORAL. Personally, I was shocked.

[hck]

Can you make the reviews public as well ? I'd love to know what the reviewers said.

[whria]

The main problem is the culture where only the first author writes all manuscript and others do almost nothing. If co-authors contributed to the revision of the manuscript, this ridiculous result can never be obtained.

[김빈남]

1st author: Plagiarist
Corresponding author and co-authors: Freeriding on Plagiarism