This sounds like a terrible story with no winners.
If it is true, it seems like a terrible case of student abuse, as well as fraud and corruption in the ISCA review process (ISCA presumably still being the ostensible top computer architecture conference), and another highly damning indictment of graduate education and universities ignoring the mental health and safety of their students.
I am also feeling less enthusiastic about the relentless drive to publish at top conferences or (sometimes literally) perish. Conference papers are often rushed to completion and full of mistakes or questionable results, and it's also hard to review them adequately given time constraints and minimal compensation or incentive for high-quality reviews. I think CS and EE may need to de-emphasize conferences and try to improve journals.
It is well-known that the review process (and not just for ISCA) is simply broken and seems to be no better than random, but I had no idea that corruption played a part in it.
> I think CS and EE may need to de-emphasize conferences and try to improve journals.
I agree. Both because it may lead to a more reasonable working pace (we have the Internet for low-latency dissemination anyway), but also because the relentless long-distance travel is not environmentally sustainable.
Journal reviews can be stacked with friends just as easily. In fact, most top CS/EE journals ask authors to recommend reviewers. (In theory, the editors will make sure the recommendations are conflict-free and sensible, but given their workloads, I wouldn't be surprised if at least some of them took the authors' suggestions at face value.)
I think computer architecture is especially prone to this sort of nonsense. The whole field has a weird insider-bias and most program committee members seem to come from the same schools, have the same advisors, and write the same sort of papers.
I find it kinda interesting that Moritz Lipp and Daniel Gruss have revolutionized our understanding of microarchitectural side channels and not a single one of their papers has appeared in ISCA/ASPLOS/MICRO. And yet, obviously flawed nonsense supposedly addressing side channels regularly manages to appear in these conferences.
I've found myself in similar situation involved with evil forces within academia (which the corrupt system creates). Just get out of institutional/ bureaucratic academia and research. It will either make (you equally evil) or break you.
I'm happy to read Huixiang's text, in that they are clear in explaining his PIs transgressions. I'm in a similar situation where my PI is threatening to kick me out of the lab or fail me after 5 years in the program after bringing him data better than all of his last 5 students combined. So I definitely know where the ECE candidate is coming from.
But the PhD candidate needed to step away from his data/project and instead familiarize himself with his options. Or befriend someone who could. This is really where hyper specialization breaks down, in that you get the "free rider" problem of people abusing the process (e.g. congressman) to parasite and bastardize others work. Now the PhD student has to become a master in a second subject (the law) to preserve the huge investment in the first subject (science). Needless to say, this ends badly a lot of the time.
It's worth pointing out that as a foreigner, he was even more isolated than an American student. Because I myself have received stellar advice from graduate student Union professionals from a university in a state far away. This student likely has several barriers, cultural and others, keeping him from effeciently defending his rights. Think about this the next time you hear about academia's obsession with foreign students
Some Universities have an international student academic dean for grad students. There will also be a Dean of graduate students too. You might ask them for help. Each department is also supposed to have a chair of graduate student advising as well, so if that isn't your PI you might ask them. There is no shame in asking for help.
Things are not that simple, I learned it the hard way:
Professors are given too much power of treating PhD students in their labs. They can cut your financial support, kick you out of PhD program (after 4,5,6 years) with some good reasons 'unproductive performance'. Deans usually don't want to get deep involved into such issues. From their perspectives, other professors are their coworkers, and will be seen for a long time. While PhD students just come and go, for people with less integrity, PhD students don't matter that much. As long as the professor doesn't cross the borderline, they are totally fine.
Life for a international PhD student is even worse. They forsake their home country life, families and friends to pursue this academic research path. They cannot easily just quit and find a job in U.S. due to the visa issue. (OPT time has a strict timing requirement for application, for most PhD students, once getting their master degree, they will lose the eligibility of OPT application)
Big tips for international PhD students, don't get your master degree just because you gain enough credits. Save it when you need an exit plan.
After 3-4 years, the sunk cost and difficult visa status can make the international students get stuck there.
The problem is that “stepping away” can alone have consequences - for international students the potential downside is massive. I vaguely recall that for some people the visa is terminated essentially the moment that they cease being a student - would you want to be a foreigner in America with an expired visa today? Maybe it’s fine for most, but what if you were a Mexican student - that clearly puts you in line for a concentration camp, would you risk it?
And this kid had already done everything he was meant to, why would they think anything else would work?
As a fellow PhD candidate, I can confirm that this sort of thing happens pretty regularly. Basically his advisor wouldn't take the risk of making the claim so he made his student do it, then put the onus on the student of "clarifying" this claim. If the claim works, they share credit. If it does not work, the blame is solely on the student for making a bad claim. If this is a seminal work, then it puts the student in a terrible situation if they want to pursue this line of work.
12 comments
[ 4.6 ms ] story [ 40.6 ms ] threadIf it is true, it seems like a terrible case of student abuse, as well as fraud and corruption in the ISCA review process (ISCA presumably still being the ostensible top computer architecture conference), and another highly damning indictment of graduate education and universities ignoring the mental health and safety of their students.
I am also feeling less enthusiastic about the relentless drive to publish at top conferences or (sometimes literally) perish. Conference papers are often rushed to completion and full of mistakes or questionable results, and it's also hard to review them adequately given time constraints and minimal compensation or incentive for high-quality reviews. I think CS and EE may need to de-emphasize conferences and try to improve journals.
It is well-known that the review process (and not just for ISCA) is simply broken and seems to be no better than random, but I had no idea that corruption played a part in it.
I agree. Both because it may lead to a more reasonable working pace (we have the Internet for low-latency dissemination anyway), but also because the relentless long-distance travel is not environmentally sustainable.
I find it kinda interesting that Moritz Lipp and Daniel Gruss have revolutionized our understanding of microarchitectural side channels and not a single one of their papers has appeared in ISCA/ASPLOS/MICRO. And yet, obviously flawed nonsense supposedly addressing side channels regularly manages to appear in these conferences.
But the PhD candidate needed to step away from his data/project and instead familiarize himself with his options. Or befriend someone who could. This is really where hyper specialization breaks down, in that you get the "free rider" problem of people abusing the process (e.g. congressman) to parasite and bastardize others work. Now the PhD student has to become a master in a second subject (the law) to preserve the huge investment in the first subject (science). Needless to say, this ends badly a lot of the time.
It's worth pointing out that as a foreigner, he was even more isolated than an American student. Because I myself have received stellar advice from graduate student Union professionals from a university in a state far away. This student likely has several barriers, cultural and others, keeping him from effeciently defending his rights. Think about this the next time you hear about academia's obsession with foreign students
Professors are given too much power of treating PhD students in their labs. They can cut your financial support, kick you out of PhD program (after 4,5,6 years) with some good reasons 'unproductive performance'. Deans usually don't want to get deep involved into such issues. From their perspectives, other professors are their coworkers, and will be seen for a long time. While PhD students just come and go, for people with less integrity, PhD students don't matter that much. As long as the professor doesn't cross the borderline, they are totally fine.
Life for a international PhD student is even worse. They forsake their home country life, families and friends to pursue this academic research path. They cannot easily just quit and find a job in U.S. due to the visa issue. (OPT time has a strict timing requirement for application, for most PhD students, once getting their master degree, they will lose the eligibility of OPT application)
Big tips for international PhD students, don't get your master degree just because you gain enough credits. Save it when you need an exit plan.
After 3-4 years, the sunk cost and difficult visa status can make the international students get stuck there.
And this kid had already done everything he was meant to, why would they think anything else would work?