Search
Without JavaScript, every record is listed below, and your browser's find command searches this page.
Lessons
- Choose a question worth answering
- Build the method and its checks together
- Use a council of models
- Start with a known result and keep an independent check
- Keep enough of the work to resume, audit and undo it
- Routine research work is a useful place to start
Accounts
- John H Boyer, philosophy
- Thomas D. Grant, law
- Exploring a question in knot theory (Robert Lipshitz), mathematics
- Tracing summaries to a radio archive (Dominik Lukeš), digital archiving
- Using physics tools to study a forest (Matthew Schwartz), physics
- Chris Blattman, economics
- William Baude, law
- Lawrence Evalyn, literary studies
- Simon Goldstein, philosophy
- Terence Tao, mathematics
- Gus Skorburg, philosophy
- Checking an R port against Stata (Francis DiTraglia), economics
- Ke Xiao, Han Li, Teng Zhang, Yangchen Xu, Runze Mao, Zhi X. Chen, engineering
- Robert Lipshitz, mathematics
- Pedro H. C. Sant'Anna, economics
- Aaron Tay, library science
- Scott Cunningham, economics
- Adam Johnson, medicine
- Joshua Sternfeld, history
- Jia Yu, Weiwei Yu, Pengfei Xiao, Fukun Xing, linguistics
- Seyed Yahya Shirazi, neuroscience
- Paul Goldsmith-Pinkham, economics
- Cameron Blevins, history
- Siddharth Mishra-Sharma, physics
- Matthew Schwartz, physics
- Max Zimmer, Nico Pelleriti, Christophe Roux, Sebastian Pokutta, computer science
- Seth Lazar, philosophy
- Zeke Hausfather, climate science
- Sam Illingworth, higher education
- Benjamin J. Shields, chemistry
- Rob Patro, genomics
- Tom Pepinsky, political science
- Rayed AlGhamdi, information systems
- Shawn Graham, history
- Tim Dettmers, computer science
- Andy Hall, political science
- Koen van Gilst, digital humanities
- Kevin Yang, field not stated
- José R. Penadés, microbiology
- Matthias Mittner, psychology
- Tiago Ventura, political science
Replication attempts
- LLM workflow run over 4,452 economics replication packages
- Coding agents replicating four scientific machine learning papers with an evidence workflow
- LLM reanalysis of 180 social and behavioural science studies
- SocSci-Repro-Bench: Claude Code and Codex reproducing 54 social-science papers
- Reproducing 48 social-science papers from the methods description alone
- Institute for Replication AI replication games: human-only, AI-assisted and AI-led teams
- AI-assisted full-paper replication of 384 political science studies
- ReplicatorBench: agents replicating social and behavioural science claims with new data
- Scott Cunningham: Claude Code replicating a PNAS text-classification paper with GPT-4o-mini
- Auditing three Kosmos radiation-biology claims against null models
- Kosmos reproducing unpublished and preprinted findings (developer-run)
- ReplicationBench: agents replicating astrophysics papers
- Analytic flexibility in silicon samples: re-examining Argyle et al. Study 3
- REPRO-Bench: agents assessing reproducibility of 112 social-science papers
- GPT-4o agent team reproducing five Alzheimer's disease studies
- PaperBench: agents replicating 20 ICML 2024 papers from scratch
- SciReplicate-Bench: reproducing algorithms from 36 NLP papers
- Independent evaluation of Sakana's AI Scientist
- CORE-Bench: computational reproducibility from code and data, and its later 'solved' status
- LLMs as simulated participants: replicating 156 psychology and management experiments