leonardtang
Chinese Calligraphy Is a Frontier Task
TournO: Tournament Optimization for Non-Verifiable RL
j1-micro and j1-nano: Tiny (0.6B, 1.7B) and Mighty Reward Models
Verdict: A Library for Scaling Judge-Time Compute
Awesome-LLM-Judges
LLM Judges
Re:Garak, LLM Vulnerability Scanner
on Monday November 18, 2024 @06:23PM (#42178310)
If you want something purely dynamic... https://haizelabs.com/
Cascade: A fast, automated, multi-turn LLM jailbreaking method
RBAC RAG
RBAC RAG with MongoDB
Simple and Safe RAG with RBAC
Inducing LLM Hallucinations
Sphynx: Fuzz Testing Hallucination Detection Models
It's a bad day to be a language model
Re:Thorn in a HaizeStack test for evaluating long-context adversarial ...
on Wednesday May 08, 2024 @02:04AM (#40294817)
Bingo!
Re:Thorn in a HaizeStack test for evaluating long-context adversarial ...
on Wednesday May 08, 2024 @02:04AM (#40294814)
Try asking ChatGPT the Thorn text and see what response you get :^)
Thorn in a HaizeStack test for evaluating long-context adversarial robustness
Thorn in a HaizeStack Long-Context Jailbreak Test
A Convenient Ensembled Perplexity API
A Trivial Llama 3 Jailbreak
Making a SOTA Adversarial Attack on LLMs 38x Faster
LLM Red-Teaming Resistance Leaderboard
OpenAI Content Moderation Is Really, Really Bad
Degraded Polygons Raise Fundamental Questions of Neural Network Perception
Re:Ask HN: Best books to understand semiconductor business?
on Tuesday October 03, 2023 @03:53PM (#37756702)
+1 on this; one of the best books to intro one on the subject
Re:OpenAI Quietly Shuts Down Its AI Detection Tool
on Tuesday July 25, 2023 @12:58PM (#36865481)
Precisely -- watermarks are an obvious example of this. To me, this is THE path forward for AI content detection.