Unverified50% confidenceFactExact time
Integrated-gradient evidence for class probabilities can be computed exactly via finite quadrature, and the sum of word contributions matches the difference between the class probability and the uninformed baseline to 10⁻¹¹ or better.
1
Sources
50%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe correlation between token-level probabilities (logprob) output by a model and factual accuracy is not strong, and fabricated facts may receive the same high generation probability as true information.69% similarUnverified研究数据表明，在数学竞赛级别问题上，启用深度推理的模型准确率可从不足30%提升至80%以上69% similarUnverifiedThe generator LLM, probe model, and prompt verbosity together contribute less than 10% of the variance.68% similarUnverifiedAssuming the data follows a Gaussian distribution, maximizing the likelihood is equivalent to minimizing the mean squared error.68% similarUnverifiedThe authors note that this significance may be affected by the small sample size and multiple comparisons, and could be due to chance.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134214API
curl https://kongchang.com/api/v1/knowledge/claims/1134214MCP
get_claim(id=1134214)