Unverified50% confidenceFactExact time
Uncensoring primarily affects a model's tendency to refuse, rather than its fundamental reasoning and generation capabilities, so capability loss is not significant for alignment-insensitive tasks such as coding.
1
Sources
50%
Confidence
Long-term
Relevance
9/24/2026
First Seen
Sources
Related Claims
UnverifiedOver-alignment is essentially a product of bias in the alignment objective function, where the model has learned to avoid criticism rather than to be genuinely helpful.78% similarUnverifiedModel abstention can avoid confident errors caused by compression distortion.76% similarUnverified负面提示（Negative Prompting）通过明确告知模型不要做什么的排除性约束，往往比正面风格描述更精确可控76% similarUnverifiedWhen a model refuses to answer certain types of questions or gives responses with a particular slant on certain topics, behind it lies a value choice made by developers, not an objectively neutral human consensus.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/945895API
curl https://kongchang.com/api/v1/knowledge/claims/945895MCP
get_claim(id=945895)