Unverified50% confidenceOpinionExact time
Benchmark scores mainly reflect performance on specific tasks and can hardly capture the impact of a model on learners' cognition, motivation and interpersonal relationships in real educational scenarios.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe effect of few-shot prompting shows strong task dependence, with task nature being the key variable determining whether examples are effective.74% similarUnverified在目标检测任务中,标注矛盾会同时影响分类分支和回归分支两个学习目标74% similarVerifiedGoodhart's Law states that when a metric becomes a target in itself it is no longer a good metric, and introducing test sets into training data causes a model's exam scores to decouple from its true capabilities69% similarUnverifiedThe objectives learned by models during training often deviate from the values and behavioral norms that humans truly expect69% similarVerifiedThe shared limitation of Hebbian learning and competitive learning is the lack of an explicit optimization objective, making end-to-end fine-tuning for downstream tasks difficult, and they are typically used for feature extraction and representation learning.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1181939API
curl https://kongchang.com/api/v1/knowledge/claims/1181939MCP
get_claim(id=1181939)