Unverified50% confidenceBenchmarkExact time

在队列准入控制任务中，RL 策略得分从基线的 −16.2 提升至 25.6

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

7/15/2026

First Seen

Valid until: 10/13/2026

Sources

decisionrl：专为运营决策打造的强化学习开源库

Related Claims

Unverified在队列准入控制任务中RL策略从基线的−16.2提升至25.683% similar Unverified在温控任务中，学习策略从-304提升到-35.8，展现显著优势60% similar UnverifiedGLM 5.2的Diff测试分数从GLM 5.1的18分跃升至46.2分59% similar Unverified在通用智能体任务上Glimmer得分达24%，高于Qwen的17%57% similar Unverified引导模式下平均端到端准确率从28.6%提升到33.7%，歧义检测F1值从45.3%提升到64.9%56% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/514222

API

curl https://kongchang.com/api/v1/knowledge/claims/514222

MCP

get_claim(id=514222)