[KongchangAI]
Unverified50% confidenceFactExact time

The attention mechanism of large language models may be drawn by the user's specific question, the names of political entities in the context, and political commentary in the training data, thereby probabilistically overpowering the influence of the system prompt.

1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/1029202
API
curl https://kongchang.com/api/v1/knowledge/claims/1029202
MCP
get_claim(id=1029202)