Unverified50% confidenceTradeoffExact time
An inherent limitation of the synthetic data framework is the gap between artificially constructed data and real-world complexity.
1
Sources
50%
Confidence
Long-term
Relevance
9/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSim-to-Real Gap refers to the systematic distributional difference between data generated in simulated environments and data collected in the real world, and is the core bottleneck in synthetic data applications74% similarUnverified合成数据可能出现模式坍塌导致多样性不足、遗漏尾部分布中稀有模式、在高维数据中引入不真实相关性等质量问题70% similarUnverifiedMost existing query reformulation evaluation datasets are either limited in scale or rely on synthetic data, creating a gap from the complexity of production environments.69% similarUnverifiedThe biggest risk of large models in data analysis scenarios is fabricating seemingly plausible but actually incorrect data when the corresponding field cannot be found.68% similarUnverifiedEarly large models in cross-source data research tasks were prone to missing key sources, numerical confusion, or broken inference chains, forcing agents to repeatedly supplement queries.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/955519API
curl https://kongchang.com/api/v1/knowledge/claims/955519MCP
get_claim(id=955519)