Unverified50% confidenceTradeoffExact time
Manual evaluation is close to the user's actual experience, but has two shortcomings: instability (subjectivity) and low efficiency.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified期望违反理论（Expectancy Violation Theory）解释了为什么即使客观上更好的输出也可能让用户感到不适，因为不符合预期本身会消耗认知资源并产生负面情绪77% similarUnverifiedUsers' perception of declining model capabilities may be influenced by rising subjective expectations, differences in task difficulty, and variation in prompt quality.72% similarUnverifiedThe two major pain points facing long-running AI agents are runaway costs and the reliability of self-evaluation.71% similarUnverified在顶会评审中Excitement低往往比Soundness低更难挽回，方法漏洞可靠补充实验修补但创新性不足难以通过Rebuttal扭转71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/915489API
curl https://kongchang.com/api/v1/knowledge/claims/915489MCP
get_claim(id=915489)