Unverified50% confidenceFactExact time
Mainstream evaluation frameworks include MMLU (breadth of knowledge across 57 disciplines), HumanEval and MBPP (code generation), GSM8K and MATH (mathematical reasoning), and MT-Bench and Chatbot Arena (human preference voting).
1
Sources
50%
Confidence
Long-term
Relevance
8/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCurrent mainstream LLM evaluation benchmarks such as MMLU, HumanEval, and MT-Bench focus primarily on knowledge capabilities and instruction-following abilities, and systematic evaluation of sycophantic behavior has yet to form an industry standard.76% similarUnverified众包式盲测评测被认为比传统学术基准测试（如MMLU、HumanEval）更能反映模型在开放式对话中的综合表现62% similarUnverifiedLMSYS 的 Chatbot Arena（lmarena.ai）以双盲对比方式让用户在不知道模型身份的情况下评分，排名结果相比商业宣传更具参考价值58% similarUnverifiedBraintrust、Patronus等偏评测视角的工具关注模型输出的质量评分58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/824049API
curl https://kongchang.com/api/v1/knowledge/claims/824049MCP
get_claim(id=824049)