Unverified50% confidenceOpinionExact time
The cautious camp believes that LLMs are good at interpolation (drawing analogies within the training distribution) but poor at genuine extrapolation (creative reasoning when faced with entirely new structures).
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified不同厂商的LLM具有不同的预训练语料、RLHF偏好对齐策略和架构设计，其错误模式的相关性更低，理论上集成效益更为显著72% similarUnverifiedIf evaluation benchmarks are highly aligned with training data, surpassing them reflects overfitting-style specialization rather than genuine capability improvement.72% similarUnverifiedCompared to MoE architectures, dense models have more uniform gradient flow during training, are less prone to expert collapse, and exhibit more deterministic inference behavior.71% similarUnverifiedDuring the pre-training phase, LLMs optimize token prediction probability through maximum likelihood estimation, lacking an inherent mechanism to 'know what they don't know,' and there is no necessary correlation between confidence and accuracy.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/913907API
curl https://kongchang.com/api/v1/knowledge/claims/913907MCP
get_claim(id=913907)