RoBERTa 기반 Reward 설계를 통한 AI Detector TPR 87.88% 감소
Notes on adversarial paraphrasing: a paper review
논리적 구문에서 거대 Matrix Weights 기반의 확률적 구조로의 패러다임 전환
Strange Truths from the Architecture of AI
Reward-based Gradient Descent를 통한 NN 행동 최적화 구현
Understanding Reinforcement Learning with Neural Networks Part 6: Completing the Reinforcement Learning Process
70ms Local NLI Judge를 통한 DSPy Reward Scoring 가속화
A 70ms Local NLI Judge Hits 0.596 Pearson r With Groq Llama 3.3 70B on DSPy Reward Scoring