Verified70% confidenceFactExact time
Large language models have a miscalibration problem, often answering in a confident tone regardless of whether the answer is correct or incorrect.
4
Sources
70%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
元认知反馈：用强化学习让大模型准确表达不确定性
hackernewshackernews7/7/2026
Related Claims
UnverifiedMathematical arguments generated by large language models are highly prone to appearing correct while actually being erroneous, and finding the errors requires word-by-word review by professional mathematicians.79% similarUnverifiedPerplexity能够提供带引用来源的答案，部分解决了大语言模型的幻觉问题79% similarUnverifiedTranslations generated by large language models read fluently even when they drop clauses, lose negations, round numbers, or invent wording that changes the meaning, which is more dangerous than the obvious errors of statistical machine translation because reviewers are more likely to overlook errors in fluent text (automation bias/fluency heuristic).78% similarUnverifiedThe study finds that these top-tier large language models frequently produce basic grammatical and semantic errors in Urdu.78% similarUnverifiedOpenAI explained in a September 2025 paper, "Why Language Models Hallucinate", that most training and benchmarks only score answers as right or wrong and give no credit for "I don't know", which leads models to lean toward guessing.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/279888API
curl https://kongchang.com/api/v1/knowledge/claims/279888MCP
get_claim(id=279888)