Unverified50% confidenceOpinionExact time
The confidence level of a language model and the reliability of its judgments are often not correlated, posing the risk of hallucinatory overconfidence.
1
Sources
50%
Confidence
Long-term
Relevance
9/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCurrent mainstream large language models exhibit significant deficiencies in both confidence calibration and selective abstention, tend to be overconfident, and when faced with questions outside the training distribution, tend to produce fluent but incorrect answers.79% similarUnverifiedWhen a model lacks accurate knowledge of a particular library, it tends to generate syntactically correct but semantically incorrect code—so-called 'confident errors.'78% similarUnverifiedLanguage models tend toward a self-confirmation bias, which limits the effectiveness of single-agent self-critique.73% similarUnverifiedAutoregressive language models have never seen 'I don't know' as a high-frequency reasonable output during training, so they are more likely to generate fabricated content at the boundaries of their knowledge than to express uncertainty69% similarUnverifiedModels lacking self-verification tend to fall into a cycle of confident errors, making it difficult to reliably complete multi-step end-to-end tasks even when initial generation accuracy is relatively high.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/955497API
curl https://kongchang.com/api/v1/knowledge/claims/955497MCP
get_claim(id=955497)