The perils of under- vs over-sculpting AGI desires
"奖励函数" 相关文章
Towards Improving Reward Design in RL: A Reward Alignment Metric for RL Practitioners
Application of LLM Guided Reinforcement Learning in Formation Control with Collision Avoidance
Misalignment from Treating Means as Ends
A Review of Reward Functions for Reinforcement Learning in the context of Autonomous Driving
我们找到3位大学教授，聊了聊越来越严重的AI幻觉
研究人员提出因果贝尔曼方程，在特定线上学习算法中可更快得到最优智能体
Improving Trust Estimation in Human-Robot Collaboration Using Beta Reputation at Fine-grained Timescales
Feature-Based vs. GAN-Based Learning from Demonstrations: When and Why
极客说｜强化学习（RL）与有监督微调（SFT）的选择以及奖励函数的优化
从RLHF、PPO到GRPO再训练推理模型，这是你需要的强化学习入门指南
如果人生是一场强化学习，你的奖励函数是什么？
The Theoretical Reward Learning Research Agenda: Introduction and Motivation
SFT并非必需！推理模型仅靠RL就能获得长思维链能力，清华CMU团队破解黑盒
Reward Hacking in Reinforcement Learning
Policy Gradient Algorithms
Anthropic: ↩️ We find that models generalize, without explicit training, from easily-discoverable dishonest strategies like sycophancy to more conce...
Anthropic: ↩️ We designed a curriculum of increasingly complex environments with misspecified reward functions. Early on, AIs discover dishonest str...
