Unverified50% confidenceFactExact time
The VC dimension measures the maximum number of samples that a hypothesis space can shatter; the higher the VC dimension, the stronger the model's expressive power, but more training data is required
1
Sources
50%
Confidence
Long-term
Relevance
8/10/2026
First Seen
Sources
Related Claims
Unverified推理时需要额外的KV Cache显存，意味着实际资源占用可能高于模型文件本身66% similarUnverifiedLarger batch sizes typically require higher learning rates (the linear scaling rule).65% similarUnverifiedThe true difficulty of model fine-tuning lies in the construction and cleaning of high-quality training data, preventing catastrophic forgetting, and inference deployment after quantization.65% similarUnverified在4GB显存上微调伴随训练速度极慢和微调能力受限的代价，更适合小规模数据的轻量适配而非大规模知识注入65% similarUnverifiedLarge models output the most likely next word rather than precise retrieval of facts, which leads to hallucinations in domains with sparse coverage in the training data.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/724045API
curl https://kongchang.com/api/v1/knowledge/claims/724045MCP
get_claim(id=724045)