Unverified50% confidenceBenchmarkExact time
Tests on five models show that whole-question accuracy on the Reproduce task ranges from 49.21% to 68.25%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/26/2026
First Seen
Valid until: 11/9/2026
Sources
Related Entities
Related Claims
Unverified斯坦福大学的AgentBench基准测试（2023年）量化证明，在复杂多步骤任务中纯ReAct范式的成功率往往不超过30%76% similarUnverifiedIn the 240-question test, ReAct's accuracy was 59.6%, Plan-execute's 61.7%, Supervisor's 65.0%, and Draft-verify's 63.7%.76% similarUnverifiedResearch shows that, for the same task, an optimized prompt can improve model output quality by 30%–50% compared to arbitrary input.74% similarUnverifiedThe accuracy of the seven models tested on PhysMent ranges from 25% to 67%73% similarUnverifiedThe benchmark produces two numbers for each model that are never merged: a task score and a false-confidence rate (the proportion of cases that should have been escalated but were answered anyway).72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1098823API
curl https://kongchang.com/api/v1/knowledge/claims/1098823MCP
get_claim(id=1098823)