Unverified75% confidenceFactTime unknown
人类在被提问时比主动回忆时能提供更准确、更完整的信息
1
Sources
75%
Confidence
Long-term
Relevance
6/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverified上下文工程比提示词工程更关注信息的长期沉淀与动态注入，而非单次对话技巧71% similarUnverifiedMemory extractors tend to have higher recognition rates for explicitly declared information, while being prone to missing or misjudging user preferences or habitual patterns that are implicit in context.68% similarUnverifiedHuman annotators exhibit a cognitive bias toward more information being better when evaluating answers, causing reward models to implicitly link length with quality, which is one of the causes of model verbosity.67% similarUnverifiedResearch shows that human evaluators in RLHF tend to prefer longer, more detailed, and friendlier-toned responses, even when those responses are not superior in factual accuracy.67% similarUnverifiedYale University research indicates that people who can easily access information tend to overestimate their actual level of knowledge, confusing 'knowing where to find an answer' with 'truly understanding the answer'.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/37583API
curl https://kongchang.com/api/v1/knowledge/claims/37583MCP
get_claim(id=37583)