This neglects the classic paper published in The Annals of Improbable Research, The Effects of Peanut Butter on the Rotation of the Earth: <a href="https://improbable.com/airchives/classical/articles/peanut_butter_rotation.html" rel="nofollow noreferrer">https://improbable.com/airchives/classical/articles/peanut_b...</a>
Not a paper, but<p>2×3×5×7×11×13 + 1 = 59×509<p>is a short counter example to the widespread misconception that adding one to the product of the first n consecutive prime numbers always yields a prime number.<p>The reason you get away with this in the infinitely-many-prime-numbers proof is that the new number may not be prime, but can be written as a product of primes that are distinct from the first n primes. Thus you still generate new prime numbers with this technique.
Though not quite as short, the Watson and Crick paper is another famously short paper:<p><a href="https://dosequis.colorado.edu/Courses/MethodsLogic/papers/WatsonCrick1953.pdf" rel="nofollow noreferrer">https://dosequis.colorado.edu/Courses/MethodsLogic/papers/Wa...</a>
Missing my favourite zero-word paper:<p>"Can a good philosophical contribution be made just by asking a question?"
<a href="https://doi.org/10.1111/meta.12599" rel="nofollow noreferrer">https://doi.org/10.1111/meta.12599</a>
When I saw this paper about “Royalactin” in Nature [1], it was fascinating subject matter for one but I was impressed that it was a single author non theoretical original research article in a high impact journal. I thought it was baller. Have been trying to search for anything similar published in recent times and have come across empty. I feel it’s emblematic how impossible it is to make any great biology breakthrough nowadays as a lone wolf. But one can aspire!<p>The guy who published it seems kooky as well. Would love to interview him some day!<p>1. <a href="https://www.nature.com/articles/nature10093" rel="nofollow noreferrer">https://www.nature.com/articles/nature10093</a>
Vaguely related but I remember the tale of an author on vacation after publication of his latest novel. He telegramed his publisher to enquire about reception:<p>“?”<p>Publisher replied:<p>“!”
Along with short papers, there are short titles. I. I. Rabi tried to get a paper published in a German journal with a one-word title:<p>"Molekularstrahlenablenkungsmethode"<p>The journal turned it down.<p><a href="https://www.aip.org/history-programs/niels-bohr-library/oral-histories/4836" rel="nofollow noreferrer">https://www.aip.org/history-programs/niels-bohr-library/oral...</a>
In computational geometry, Raimund Seidel wrote a paper that proves an upper bound theorem for polytopes in two sentences in the abstract. The rest of the paper just comments on the result.<p><a href="https://www.sciencedirect.com/science/article/pii/092577219500013Y?via%3Dihub" rel="nofollow noreferrer">https://www.sciencedirect.com/science/article/pii/0925772195...</a>
My high school math teacher told as about how students at his university competed for the shortest bachelor paper.<p>Not getting a paper published is par for the course, but having to retake your bachelor examination is quite the hassle. The risk and associated bragging rights seemed quite big.
A recent 5-page cryptography paper ended up winning a best paper award (albeit for an extended version)<p><a href="https://eprint.iacr.org/archive/2020/945/1596227165.pdf" rel="nofollow noreferrer">https://eprint.iacr.org/archive/2020/945/1596227165.pdf</a>
It's missing the shortest meaningful paper:<p>Fiengo, Robert, and Howard Lasnik. 1972. “On Nonrecoverable Deletion in Syntax.” Linguistic Inquiry 3 (4): 528. <a href="https://i.imgur.com/vLntfCp.jpg" rel="nofollow noreferrer">https://i.imgur.com/vLntfCp.jpg</a>
Obliquely related, the shortest paper title I am aware of: H=W. By Norman G. Meyers and James Serrin, Proceedings of the National Academy of Science 51 (1964), 1055–1056.
Short abstracts such as ‘yes’, ‘no’, &c of papers whose titles are questions are amusing but not as helpful as those whose titles straightforwardly state the answer (e.g., ‘X is Y’ is preferable to ‘Is X Y?’ with abstract ‘yes’).
Brian Josephsons' paper on the superconducting tunneling effect that's now named after him was one of the first (and last) papers he ever published and netted him a Nobel prize. It's just 2 pages long I think [1]. Can't be sure as I cannot get the full text since it's paywalled, pretty sure it's only 2-3 pages though.<p>1: <a href="https://www.sciencedirect.com/science/article/abs/pii/0031916362913690?via%3Dihub" rel="nofollow noreferrer">https://www.sciencedirect.com/science/article/abs/pii/003191...</a>
Hmm. I'm wondering how the short lengths of these papers might cause them to have lower information-theoretic entropy than the (in)famous "chicken" talk[1]?<p>[1] <a href="https://www.youtube.com/watch?v=yL_-1d9OSdk">https://www.youtube.com/watch?v=yL_-1d9OSdk</a>
As a rule, when trying to convey information, I try to write and speak plainly. I try to avoid jargon.<p>I have found many academic papers in the faux sciences to be extremely dense and full of terms that are only known to the priests of that arcane subject (still subsidized by taxes as if the result is a common good).<p>If you have a point, say it. There is no need to write in legalese. When I see supposed research written like this, I assume it’s a grift written just for the tiny group of academics tenured in that subject, who review each others’ papers every year, buy each others’ books, and keep the perpetual motion machine of funding running until they hit retirement.
I was hoping that LLMs would help with "writer's block", but their limited context window has been a PITA. Has anyone have had any success with ROPE scaled models or claude 100k?