Hacker News Reader

Comment by tornikeo | Hacker News Reader

> First-author paper published at a top conference

I now find that requirement in most AI-related high-tech jobs. Starter salary for these jobs is often $150k+. When someone is willing to pay you $150k+ for having published a paper, fraud definitely makes (financial) sense. Basically, the problem is the demand, and the demand corrupts the metrics (h-index).

I think that having a paper published would become less and less significant in the future. With time, businesses will also move onto other signals of success.

View on HN

It's not just a way for landing a job either. It's also a ticket for white collar immigration on "specialist" visas that have a much shorter waiting list, higher quota or better perks than regular work-based visas. I was blown away the first time I heard people talking openly about this in my field - it was considered totally normal for them to pay money to have an academic portfolio constructed on their behalf specifically for the purpose of having these "published papers" and "unique research" on their record so they could qualify for "specialist" visas. It adds a whole nother layer to the warped idea of good immigrants vs bad immigrants.

The papers published at top conferences are not the papers that is coming from this "industry" as the paper calls it.

These fraudulent papers are identified like this:

> For instance, of the 79 papers that one editor had handled at PLOS ONE, 49 have been retracted.

That's not what's happen at top conferences.

It is well-known that top-conferences had and still have many problems. Some examples: There used to be the problem with authors adding new co-authors after acceptance, aka "selling seats". There is a debate about how many papers one should be allowed to submit, as some people with money and influence are heavily franchising. It is unclear to what extent there is implicit and explicit reviewer collusion. Even double-blind reviews don't really solve the problem.

If we don't admit that there are fundamental problems that affect all of us (academics) and instead pretend it is only the lesser people who f things up, we'll all be screwed sooner or later.

> There used to be the problem with authors adding new co-authors after acceptance, aka "selling seats".

Top AI conferences allow that? That's insane.

They do not allow this any more. So _that_ problem has actually been addressed.

People are waking up to this, I have seen companies asking for paper reproduction on the spot before handing out offers.

Reply on news.ycombinator.com