Unverified50% confidenceOpinionExact time
Most general-purpose large models suffer from overconfidence issues, where their output probabilities reflect token distribution preferences more than genuine uncertainty estimates.
1
Sources
50%
Confidence
Long-term
Relevance
10/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGeneral-purpose large models commonly suffer from overconfidence issues in confidence calibration, tending to output confidence levels close to 100%.79% similarUnverifiedDeep learning models are generally prone to overconfidence, where the probability values output by a model do not truly reflect the actual probability of its predictions being correct — a phenomenon known as poor calibration.76% similarUnverifiedConfidence values typically originate from the model's softmax probability distribution over each option — a large probability gap yields a high confidence value, while a uniform distribution yields a low one.76% similarUnverifiedMany models still exhibit high confidence in incorrect answers, presenting a confidence calibration problem.75% similarUnverifiedThere is an essential difference between the probability scores (logprobs) output by LLMs and calibrated confidence; large language models are widely known to be overconfident, particularly on out-of-distribution samples.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/994500API
curl https://kongchang.com/api/v1/knowledge/claims/994500MCP
get_claim(id=994500)