Unverified50% confidenceSolutionExact time
AdvRole's Rewriter is trained with a performance-gap reward: if a rewrite causes the current Actor's score to drop significantly compared to the original scenario, that rewrite receives a higher reward.
1
Sources
50%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRecursive training makes extensive use of task completion rates and code execution results as reward signals, lacking explicit supervision for narrative consistency, which may lead to catastrophic forgetting of general language modeling capabilities, and this is more pronounced in models with small parameter counts.67% similarUnverified智能体的偏好受训练数据、提示词、版本迭代等多重因素影响，今天有效的优化手段可能在下一次模型更新后失效64% similarUnverified分配给预训练的算力越多，下游任务的pass@1奖励就越高，预训练损失越低模型在后续RL中获得的收益越大64% similarUnverified逆强化学习试图从演示中推断出奖励函数，理论上更优雅但计算成本更高64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949047API
curl https://kongchang.com/api/v1/knowledge/claims/949047MCP
get_claim(id=949047)