Unverified50% confidenceFactExact time
Latent Compaction training uses only generic prompts about the document and does not access any downstream questions or answers.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEvaluating the marginal likelihood using only the last-layer weights requires no access to training data.71% similarUnverified上下文学习（ICL）方式受限于LLM的上下文窗口长度，无法将大规模训练集完整纳入提示词中63% similarUnverifiedThe classifier was trained using only supervised fine-tuning (SFT), public datasets, and two simple tricks, without relying on closed-source data or expensive custom pipelines63% similarUnverified开放权重仅公开训练完成后的神经网络参数文件，使用者可加载推理和微调，但无法复现训练过程或审计数据来源，与完全开源不同62% similarUnverified指令微调不需要从头训练，通常仅需基础训练计算量的1%以下即可显著改变模型行为模式62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1037893API
curl https://kongchang.com/api/v1/knowledge/claims/1037893MCP
get_claim(id=1037893)