Public KG record
GRPO
Recorded overview
KG-recorded mechanism related to GRPO; detailed English notes are not yet available.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Capabilities: 为部分正确页面提供部分得分
- Capabilities: 按对话、里程碑和轮次提供奖励
- Capabilities: 为工具使用提供奖励
- Capabilities: 提供对话级、里程碑级和轮次级奖励
- Capabilities: 每次尝试提供一个结果奖励
Evidence coverage
The KG currently records 7 known information dimensions and has 12 undocumented dimensions, linked to 24 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- lite.gym — uses
- TRL — uses
- Agentic Reinforcement Learning — uses
- EnvHarness — supports
- LFM2 — applied to