GRPO
Group Relative Policy Optimization (GRPO) is a reinforcement learning algorithm for fine-tuning large language models that eliminates the separate critic (value) network used by PPO
Chinese AI
Explore Chinese AI through related topics and the articles other pages reference most.
Explore articles
Browse subtopics: Reinforcement Learning
Articles that also belong to these categories. Counts cover all of Chinese AI.
All articles (287)Large Language Models (91)Open Source AI (76)AI Companies (65)AI Models (61)AI Hardware (37)Mixture of Experts (28)Multimodal AI (27)Robotics (27)Generative AI (22)Humanoid Robots (22)Video Generation (21)People (18)AI Policy & Regulation (16)Computer Vision (16)Reasoning Models (14)AI Tools & Products (11)Embodied AI (11)AI Infrastructure (10)Robotics Companies (10)AI Agents (9)AI Code Generation (9)Image Generation (8)Autonomous Vehicles (7)Research Organizations (7)Conversational AI (6)Developer Tools (5)Machine Learning (5)Robot Hardware (5)Speech & Audio AI (5)Venture Capital (4)AI Events (3)Healthcare AI (3)Information Retrieval (3)NVIDIA (3)Universities (3)AI Incidents & Controversies (2)AI Inference (2)Artificial Intelligence (2)Deep Learning (2)Enterprise AI (2)Mathematics (2)Model Architecture (2)Natural Language Processing (2)Reinforcement Learning (2)Small Language Models (2)Training & Optimization (2)Transformer Models (2)Voice AI (2)AI Ethics (1)AI History (1)AI Safety (1)AI in Gaming (1)Computer Science (1)Data & Datasets (1)Finance AI (1)World Models (1)
Showing 1-2 of 2 articles
Kimi K1.5
Kimi K1.5 is a multimodal reasoning large language model developed by Moonshot AI, a Beijing-based artificial intelligence company.