Y Hacker News new | ask | show | jobs

user: leonardtang created: 2019-10-01 karma: 41 submissions: EvoForge: Evolutionary Harness Optimization 2 points | 0 comments Chinese Calligraphy Is a Frontier Task 1 points | 0 comments TournO: Tournament Optimization for Non-Verifiable RL 3 points | 0 comments j1-micro and j1-nano: Tiny (0.6B, 1.7B) and Mighty Reward Models 3 points | 0 comments Verdict: A Library for Scaling Judge-Time Compute 3 points | 0 comments Awesome-LLM-Judges 2 points | 0 comments LLM Judges 2 points | 0 comments 0 points | 0 comments Cascade: A fast, automated, multi-turn LLM jailbreaking method 2 points | 0 comments RBAC RAG 1 points | 0 comments RBAC RAG with MongoDB 2 points | 0 comments Simple and Safe RAG with RBAC 2 points | 0 comments Inducing LLM Hallucinations 2 points | 0 comments Sphynx: Fuzz Testing Hallucination Detection Models 2 points | 0 comments It's a bad day to be a language model 2 points | 1 comments 0 points | 0 comments 0 points | 0 comments Thorn in a HaizeStack test for evaluating long-context adversarial robustness 19 points | 11 comments Thorn in a HaizeStack Long-Context Jailbreak Test 5 points | 0 comments A Convenient Ensembled Perplexity API 1 points | 0 comments A Trivial Llama 3 Jailbreak 70 points | 47 comments Making a SOTA Adversarial Attack on LLMs 38x Faster 2 points | 0 comments LLM Red-Teaming Resistance Leaderboard 2 points | 0 comments OpenAI Content Moderation Is Really, Really Bad 2 points | 1 comments Degraded Polygons Raise Fundamental Questions of Neural Network Perception 1 points | 0 comments 0 points | 0 comments