Verified65% confidenceFactExact time
The LLM-as-a-judge evaluation model assigns higher scores to outputs whose style is closer to its own training data, which is known as self-preference bias.
3
Sources
65%
Confidence
Long-term
Relevance
9/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe framework is highly dependent on the review quality of frontier LLMs, and biases inherent in the LLMs may be propagated into the final training data.76% similarUnverifiedSpecialized models (classifiers, rerankers, entity models) are usually easier to calibrate than an LLM's self-reported confidence, because their training objective directly optimizes the correspondence between scores and correctness.75% similarUnverified参数量相似的情况下，训练数据质量分布的差异往往比架构差异更能决定特定场景的表现上限75% similarUnverifiedIf evaluation benchmarks are highly aligned with training data, surpassing them reflects overfitting-style specialization rather than genuine capability improvement.75% similarUnverifiedMultiple studies in 2024 found that the strong performance of certain models on standard benchmarks resulted from benchmark test questions being mixed into the training data (data contamination).72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/936117API
curl https://kongchang.com/api/v1/knowledge/claims/936117MCP
get_claim(id=936117)