Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

World & Business

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

An "AI submissions" journal is reportedly in the works · Digg

An "AI submissions" journal is reportedly in the works

The user describes the journal as a project that will "parallel arXiv."

16 Sources, 24d ago, first seen 24d ago

TLDR

In a reply, a user says an "AI submissions" journal is being made and will "parallel arXiv."

Combined views

12.9K

16 Sources, first seen 24d ago

85 likes

Combined views

12.9K

16 Sources, first seen 24d ago

85 likes

10 comments

11 saves

8 reposts

10 comments

11 saves

8 reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Today's Rank

—

Not ranked yet

Today's Rank

—

Not ranked yet

16 Sources

@yaroslavvb@tdietterich There is an "AI submissions" journal being made that will parallel arXiv. Cc @bohannon_bot24d

@Kangwook_Leesimilarly, the notion of "human-assisted" won't exist in a few years. it's either 100% human-done (hand-made papers) or 100% ai-done.23d

@ChenhaoTanA separate repository makes a lot of sense for the AI-written/AI-verified. I think emphasizing human-written in the first scenario is distracting. We should reduce the incentive to seek publication so that the first category is for people who think publishing would still make sense.23d

@DimitrisPapailRT @Kangwook_Lee: similarly, the notion of "human-assisted" won't exist in a few years. it's either 100% human-done (hand-made papers) or…23d

@adamwslave@krismicinski It’s really important that you learn and practice the skills regarding synthesis of research. Having an LLM do the research assistant bit is fine, but if you’re unable to synthesize/interpret what’s gathered, why are you in academia in the first place?22d

@zehavocRT @adamwslave: @krismicinski It’s really important that you learn and practice the skills regarding synthesis of research. Having an LLM d…22d

@anmarasovic@TuhinChakr I'm worried who checks review quality – even our meta-reviewers aren't reliable... I still think we need desk reject of anything that smells like AI slop. I'd like that an expert reviewer checks the intro of any Pangram-flagged paper and reject if they can't understand it.22d

@TuhinChakr@anmarasovic Or we develop a model that can identify scientific slop. This could be valuable if combined with AI detection22d

@aran_nayebiarXiv should remain a repository of timestamped results, and not serve the job of an editor/gatekeeper. This is still an incredibly valuable service imho b/c at this rate, only an LLM reviewer can keep up with the scale of submissions.22d

@delipraoalso, if students produce increasingly larger volume of output due to agents, the due diligence workload on the PI becomes hard to handle and error prone. one solution is to pair junior students with senior students, reinventing middle managers, but that also means you have a steady supply of students and grant money to have a pipeline of students at different experience levels.21d

16 Sources

@yaroslavvb@tdietterich There is an "AI submissions" journal being made that will parallel arXiv. Cc @bohannon_bot24d

@Kangwook_Leesimilarly, the notion of "human-assisted" won't exist in a few years. it's either 100% human-done (hand-made papers) or 100% ai-done.23d

@ChenhaoTanA separate repository makes a lot of sense for the AI-written/AI-verified. I think emphasizing human-written in the first scenario is distracting. We should reduce the incentive to seek publication so that the first category is for people who think publishing would still make sense.23d

@DimitrisPapailRT @Kangwook_Lee: similarly, the notion of "human-assisted" won't exist in a few years. it's either 100% human-done (hand-made papers) or…23d

@adamwslave@krismicinski It’s really important that you learn and practice the skills regarding synthesis of research. Having an LLM do the research assistant bit is fine, but if you’re unable to synthesize/interpret what’s gathered, why are you in academia in the first place?22d

@zehavocRT @adamwslave: @krismicinski It’s really important that you learn and practice the skills regarding synthesis of research. Having an LLM d…22d

@anmarasovic@TuhinChakr I'm worried who checks review quality – even our meta-reviewers aren't reliable... I still think we need desk reject of anything that smells like AI slop. I'd like that an expert reviewer checks the intro of any Pangram-flagged paper and reject if they can't understand it.22d

@TuhinChakr@anmarasovic Or we develop a model that can identify scientific slop. This could be valuable if combined with AI detection22d

@aran_nayebiarXiv should remain a repository of timestamped results, and not serve the job of an editor/gatekeeper. This is still an incredibly valuable service imho b/c at this rate, only an LLM reviewer can keep up with the scale of submissions.22d

@delipraoalso, if students produce increasingly larger volume of output due to agents, the due diligence workload on the PI becomes hard to handle and error prone. one solution is to pair junior students with senior students, reinventing middle managers, but that also means you have a steady supply of students and grant money to have a pipeline of students at different experience levels.21d