Unverified50% confidenceFactExact time
During training, REMORY uses KL divergence to measure the difference in output distributions between a frozen LLM seeing the full history and seeing the summary plus residual tokens.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDuring LLM training, memorization occurs, where models may verbatim store fragments of training data and reproduce them during inference.75% similarUnverified潜在缓存技术将LLM推理中的KV Cache思想迁移至训练阶段，对轨迹历史帧预先提取潜在表示并缓存74% similarUnverifiedMLP层充当模型的长期事实记忆库，其键值对是预训练阶段固化在权重中的静态参数66% similarVerifiedThe memory of LLMs depends on the context window, which is the maximum number of tokens the model can process in a single inference.65% similarUnverified潜在缓存技术源于大语言模型中的KV Cache机制，Mistral将其从推理阶段迁移至训练阶段65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/998482API
curl https://kongchang.com/api/v1/knowledge/claims/998482MCP
get_claim(id=998482)