Unverified50% confidenceFactExact time
The most common complaint about the AI examiner in NYU Stern's Fall 2025 cohort was question stacking, corresponding to a 31% stacking-turn rate, despite the prompt explicitly stating that two questions must never be asked in the same turn.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe expert designed a self-assessment test of five questions on 'whether you talk too much,' with more than three affirmative answers indicating a high likelihood of being an over-talker.62% similarUnverifiedIn the four-bucket classification of 322 stagnation-rule injection events, only 14% of injections actually landed on reasoning chains that were genuinely stuck.60% similarUnverifiedThe sample distribution represented by the 95.8% figure may deviate from the types of vulnerabilities encountered in real-world work, as entries that are difficult to evaluate automatically tend to be high-severity vulnerabilities with complex logic and longer exploit chains.60% similarUnverified2023年多项研究证明，对同一基准采用不同措辞重新提问，部分模型表现会显著下滑60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002828API
curl https://kongchang.com/api/v1/knowledge/claims/1002828MCP
get_claim(id=1002828)