Unverified50% confidenceOpinionExact time
Jia Xiaosong believes that if training remains dominated by imitation learning, changing the architecture cannot bypass the inherent problems of imitation learning; valuable directions include filtering valuable data, going beyond brute-force imitation learning, designing more effective reward signals, and enabling models to reinforcement-learn tens of thousands of skills simultaneously.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedCompared with traditional reinforcement learning, imitation learning typically requires only tens to hundreds of demonstration trajectories for a robotic arm to learn grasping and placing, whereas reinforcement learning typically requires defining a reward function in a simulation environment and having the agent repeatedly trial and error, with sample sizes often in the millions.75% similarUnverifiedThis kinematic prior appears under both reinforcement learning and imitation learning training paradigms, possibly reflecting common knowledge related to the humanoid embodiment rather than a product specific to a particular optimization process.74% similarUnverified行为克隆预训练解决了强化学习的冷启动问题：纯强化学习在稀疏奖励环境中往往需要数亿次交互才能学到基本可用策略74% similarUnverifiedEffective reinforcement learning training requires maintaining a certain error rate, keeping the model in a range where it can answer correctly only with effort, which is consistent with the educational psychology concept of the Zone of Proximal Development73% similarVerifiedReinforcement learning training is the key stage in which a model transitions from general language capability to precise task execution, further optimizing reasoning, planning, and instruction-following through reinforcement learning after large-scale pre-training.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065563API
curl https://kongchang.com/api/v1/knowledge/claims/1065563MCP
get_claim(id=1065563)