Unverified50% confidenceOpinionExact time
Moderation model training data is heavily reliant on historical violation samples and lacks the ability to distinguish artistic, academic, and cultural contexts.
1
Sources
50%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe quality and diversity of training data directly affects the reliability of model outputs, and training data containing large amounts of low-quality, contradictory, or false information can cause models to learn incorrect patterns.73% similarVerifiedBenchmarks face contamination risks; if training data contains test questions or approximate variants, a model's high scores may partly stem from memorization rather than genuine capability.70% similarUnverifiedOnce false information is incorporated into a model's training data, it is very difficult to completely remove it from an already trained model even if it is subsequently debunked.70% similarUnverified现有训练数据审计实践主要依赖开发者自我声明（如Meta的Llama系列、Stability AI的模型卡片），缺乏独立可验证的技术手段69% similarUnverifiedTraditional large model training relies heavily on manually annotated data and external corpus collection, which are typically the bottlenecks constraining iteration speed.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/938938API
curl https://kongchang.com/api/v1/knowledge/claims/938938MCP
get_claim(id=938938)