Comments | ferventsquirrel: It was only a matter of time...

ferventsquirrel

It was only a matter of time...

Jun 25, 2009 19:32

A computer generated CS paper has been accepted to a peer-reviewed journal.

CRAP paper accepted by journal

The journals "director of publications" has the lamest excuse. Seriously.

Comments 12

rebkos June 26 2009, 03:28:44 UTC

*snerk* I computer paper written by a computer... Isn't one of the test of AI suppose to be a human interacting with an unknown (computer or human) and if they can't tell it's a computer that's good...

aepfelx June 26 2009, 03:34:13 UTC

right, except the turing test assumes there's a human on at least one side. the scigen pranks serve to point out that certain "conferences" and "journals" don't seem to have any sentient life on the other end filtering the submissions (or at least no sentient life interested in anything other than charging the authors for publication)

yellowphoton June 26 2009, 04:22:59 UTC

I dunno... isn't that the more charitable explanation? Seems to me just as likely it was a human on the other side who's brain just glazed over during the reviewing proess...

aepfelx June 26 2009, 04:37:15 UTC

uh. really, that's far too cynical. yea, not everyone takes peer review as seriously as they should but, really, this is miles past that.

let me assure you, no one that could be reasonably labeled a computer scientist has ever reviewed any of the scigen papers.

Thread 6

robynjade13 June 26 2009, 04:14:51 UTC

Hm, and I'm up to my sixth rejection. Clearly I should have had a computer write my paper.

tt6681_theresat June 26 2009, 04:26:17 UTC

Or submit to one of these journals....

aepfelx June 26 2009, 04:43:17 UTC

whoa. i applaud your persistence (and without a drop of sarcasm here, i swear)

robynjade13 June 26 2009, 04:49:40 UTC

Hah, thanks for that. Never thought of it in terms of persistence. More in the "insanity is doing the same thing repeatedly and expecting something different to happen" way. But I appreciate the thought.

Thread 5

marknn July 15 2009, 02:12:26 UTC

That's not actually the first time -- the first paper was accepted to WMSCI back in 2005. I even [tried to] read the original -- it was very strange experience. On one hand I felt that i was reading properly structured sentences, but on the other by the time I finish reading a sentence I felt that I forgot what it started with or what it was about.

Amusingly enough this one is generated by the same program. I wonder if the quality improved since then...