AI Hallucinates Entire Research Field; Peer Reviewers Impressed
In a landmark study, a large language model successfully generated 47 novel research papers, 3 academic conferences, and a Nobel Prize committee that does not exist.
Latest News
Our approach is simple, elegant, and requires 8 A100 GPUs to run.
I3E TBE (Efficiency Special Issue)
|
Dr. C. Omputationally-Free
Explaining Transformers Without Any Mathematics Whatsoever
A Survey of Negative Results Nobody Wanted to Publish
A Taxonomy of LLM Hallucinations, Illustrated With Examples From This Paper
On the Statistical Inevitability of Rejection: A 10-Year Study
Our Benchmark Is Contaminated (And Yours Probably Is Too)
I3E Trashactions on Patternless Analysis and Mostly Imaginings
I3E Trashactions on Organizational Inefficiency and Leadership
I3E Trashactions on Catastrophic P-value Shopping
I3E Trashactions on Things Nobody Told The Professor
I3E Trashactions on Reproducibility Problems in Imaginary Institutions
I3E @ Hype
Sponsored ContentCAREERS
PhD Position: Unpaid Labor with Prestigious University Branding
Postdoc Wanted: 3 Years of Experience Required for Entry-Level Pay
Senior AI Researcher: Must Have 15 Years of Experience in 5-Year-Old Technology
Assistant Professor: Teach 4 Courses, Publish 10 Papers, Raise $1M in Grants (Annual)
News Features
AI Hallucinates Entire Research Field; Peer Reviewers Impressed
In a landmark study, a large language model successfully generated 47 novel research papers, 3 academic conferences, and a Nobel Prize committee that does not exist.
Researchers Achieve Local Minimum in Both Loss Function and Mental Health
A team of 12 PhD students report converging to suboptimal solutions across all metrics after 18 months of training.
Featured Videos
[Launch Special] First Annual Lunar New Year Gala of Bottom-Tier Academic Journals
23:42
How to Write an IEEE Paper in 3 Minutes
3:14
Reviewer #2 Training Guide: From Beginner to Career-Ender
12:00
Coming Soon
Episode 1
Is Your Research Reproducible? (Spoiler: No)
42:00
Coming Soon
Episode 2
The P-Value Whisperer
38:15
Coming Soon
Episode 3
Peer Review: A Horror Anthology
55:30
Coming Soon
Episode 4
How to Survive a Conference Buffet
29:45
Coming Soon
Episode 5
Citation Needed: The Musical
47:00
How to Get Published
A simple three-step process — in theory
Step 1
Submit Your Work
Upload your manuscript, data, and the names of every reviewer who has ever wronged you.
Step 2
Survive Review
Wait 8–18 months while anonymous experts question your methodology, intelligence, and life choices.
Step 3
Achieve Publication
Celebrate briefly before immediately beginning work on the mandatory revision that will take another year.
COMMENTARY
View AllAn Ode to Reviewer #2: You Complete Me (And Then Reject Me)
Gradient Descent Into Madness: My PhD Journey
Why I Let GPT Write My Related Work Section (And You Should Too)
Peer Review Has Given Me PTSD. Here Are My 47 Rejection Letters.
Ranking Academic Conferences Solely by Buffet Quality (2025 Edition)
I Tried to Reproduce My Own Paper. It Took Three Years.
Expert opinions, hot takes, and peer-reviewed grievances — delivered straight to your inbox before they can be rejected.
Explore Commentary