Unverified50% confidenceFactExact time
UtilityAUC can capture how well a model aligns with utility preferences, not merely error frequency.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe researchers introduced the UtilityAUC metric to measure the area under the utility curve determined by the model's response ordering.72% similarUnverifiedTheoretical analysis shows that UtilityAUC can distinguish between models that perform similarly on standard accuracy and AURC.68% similarUnverifiedWhen evaluating the entire workflow, system-level metrics include: the proportion of erroneous outputs accepted by users, the survival rate of errors through the verification process, the speed of correcting persistent errors, and the proportion of supposedly independent sources that share the same ancestor.67% similarUnverified可观测性（Observability）方案从真实用户反馈和长期调用过程中采集数据，为模型迭代优化提供量化依据66% similarUnverifiedUsers can objectively verify whether a model has degraded by building a personal "regression test set"—retaining a fixed set of representative prompts to periodically evaluate model output quality.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073850API
curl https://kongchang.com/api/v1/knowledge/claims/1073850MCP
get_claim(id=1073850)