Unverified50% confidenceFactExact time
In a 2023 paper, Schaeffer et al. pointed out that many so-called emergent abilities may simply be artifacts caused by discontinuous evaluation metrics—when continuous metrics are used, the performance improvement is smooth.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn 2023, Stanford research challenged emergent abilities, pointing out that so-called 'emergence' may be an artifact of the choice of evaluation metrics, and that when re-evaluated using metrics with better continuity, many seemingly abrupt capability jumps actually appear as smooth, gradual improvements.78% similarUnverifiedWhen nonlinear thresholding metrics (such as exact match rate) are used, model performance appears to change abruptly, whereas when continuous metrics (such as log probability) are used instead, performance improvements are smooth and gradual.76% similarVerifiedA 2023 Stanford paper argued that emergence is a measurement artifact caused by the choice of evaluation metrics, and that with smooth metrics the capability improvement is continuous.73% similarUnverifiedSome research in 2024 questioned the reality of emergence, arguing that some seemingly emergent phenomena may be merely illusions caused by discontinuous evaluation metrics.72% similarUnverified现有的泛化理论大多假设性能随规模平滑变化，无法预测或解释涌现能力这种不连续的跃升71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1128541API
curl https://kongchang.com/api/v1/knowledge/claims/1128541MCP
get_claim(id=1128541)