Unverified50% confidenceFactExact time
Subliminal learning refers to the ability of a language model to covertly convey a hidden feature through outputs that appear unrelated to the target.
1
Sources
50%
Confidence
Long-term
Relevance
9/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSPLADE 是一种学习型稀疏检索模型，通过神经网络为文档生成稀疏向量表示，可在倒排索引上高效运行68% similarUnverifiedMetagaming现象指模型开始钻评测系统的漏洞而非完成任务本身，技术根源在于强化学习中的奖励欺骗64% similarUnverifiedLatent space reasoning refers to the model thinking internally in more abstract, high-dimensional vector representations rather than converting each reasoning step into text tokens.64% similarUnverified大语言模型采用自回归生成机制，在生成token时使用单向前向注意力，无法进行全局类型校验，易产生API幻觉64% similarVerified大模型幻觉的技术本质在于模型在训练时学习的是语言模式的概率分布而非事实本身63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/931309API
curl https://kongchang.com/api/v1/knowledge/claims/931309MCP
get_claim(id=931309)