curius graph

all pages

showing 23101-23150 of 170282 pages (sorted by popularity)
[2511.21654] EvilGenie: A Reward Hacking Benchmark
[2511.21654] EvilGenie: A Reward Hacking Benchmark
Indo-European Explorer: A 6,000-Year Journey
Opinion | What if Labor Becomes Unnecessary? - The New York Times
The best new novels to read this spring
The engine of Germany's wealth is blocking its future | The European Correspondent
30hr Open Weight Safety Projects - Google Docs
Why Harry Styles Loves Running: “It’s Just You and a Pair of Shoes.”
Insider Journalism - by Robin Hanson - Overcoming Bias
Statisticism: How Cluster-Thinking About Data Creates Blind Spots
Distilling Replacing Guilt — LessWrong
Moral Reality Check – Unstable Ontology
[2603.05414] Dissociating Direct Access from Inference in AI Introspection
Our Team | until
The Repugnant Conclusion (Stanford Encyclopedia of Philosophy)
🟡 Iran War continues, Strait of Hormuz remains closed, sharp drop in Chinese aircraft flying near Taiwan, Alibaba AI agent mystery || Global Risks Weekly Roundup #10/2026
Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation — AI Alignment Forum
Log In | Robinhood
[2603.05414] Dissociating Direct Access from Inference in AI Introspection
What a time to be an oncologist - by Olivia Webb Kosloff
Amy Tam on X: "When code is free, research is all that matters" / X
Partial Lean formalization of Analysis I — Verso
How To Become a Mathematical Genius - by Sinéad O’Sullivan
The Untold Chaos Behind a $3 Billion AI Startup Launch - YouTube
The First Crusade | The Salahuddin Generation | Ep. 2 | Dr. Hassan Elwan - YouTube
Versa Diary - Google Docs
The Mog Language Guide | Mog
Evidence that recurrent circuits are critical to the ventral stream’s execution of core object recognition behavior | Nature Neuroscience
Fantastic Beasts and How to Rank Them | The New Yorker
We Should Revisit Literate Programming in the Agent Era | silly business
[2510.16062] Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
[2510.16062] Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
Feedback eaters and how to spot them - by Carmen - Altered
Rohan Paul on X: "Self‑Correction Bench shows 1 word can flip 64% failure into success. Large language models often spot errors in a user prompt yet ignore identical errors in their own output. This paper measures that gap and shows a simple prompt tweak almost erases it. The authors build https://t.co/r8i7OfO5Py" / X
CorrectBench: A Benchmark of Self-Correction in LLMs
"Clean" Code, Horrible Performance - YouTube
Why Escalation Favors Iran | Foreign Affairs
Cursor's Third Era: Cloud Agents — ft. Sam Whitmore, Jonas Nelle, Cursor - YouTube
Times New Roman Turns Right - McSweeney’s Internet Tendency
How are MPs passing so many bills without voting? | CBC News
Canadian military personnel identified on white supremacist dating site | CBC Accessibility
Discover – type.lol
Kaluza–Klein theory
luck/luck.md at main · soleio/luck
justinzwu.com
The-Complete-Guide-to-Building-Skill-for-Claude.pdf
FrameBook.
Daniel Kokotajlo's Shortform — LessWrong
Compensation as a Reflection of Values / Oxide
Shreya Mogulothu