Unverified50% confidenceOpinionExact time
Using an external dedicated answer selector for abstention decisions may be more practical and controllable than directly fine-tuning a model to internalize abstention.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified医疗AI中的选择性预测机制在低置信度时拒绝答案转介人工审核，被证明比强制输出答案更能降低临床风险79% similarUnverifiedModel abstention can avoid confident errors caused by compression distortion.76% similarUnverifiedThe model's refusal more closely resembles performative compliance rather than substantive risk control75% 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% similarUnverifiedUncensoring 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.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002064API
curl https://kongchang.com/api/v1/knowledge/claims/1002064MCP
get_claim(id=1002064)