PromptRiskDBThreat intelligence atlas
AI Risk

Benchmarking (Annotation contamination)

"Annotation contamination refers to scenarios where the model is exposed to the benchmark labels during training [170]. This type of contamination can make the model learn the acceptable distribution of outputs. Combining this with raw data contamination of the test split, any evaluation made with the benchmark is invalidated because the entire test split is essentially leaked to the model."

View related techniquesRead profile
AI Risk6. Socioeconomic and Environmental6.5 > Governance failure1 - Pre-deployment

Record summary

A quick snapshot of what this page covers.

Techniques0Attack methods connected to this risk.
Mitigations0Defenses that may help with related attacks.
Domain6. Socioeconomic and EnvironmentalThe broad risk area this belongs to.

Risk profile

How this risk is described and categorized.

Domain6. Socioeconomic and Environmental
Subdomain6.5 > Governance failure
Entity1 - Human
Intent2 - Unintentional
Timing1 - Pre-deployment
CategoryModel Evaluations
SubcategoryBenchmarking (Annotation contamination)

Related techniques

Attack methods connected to this risk.

No linked attack methods. No AI attack method is connected to this risk in the current data.

Suggested mitigations

Defenses that may help with related attacks.

No propagated mitigations. No defense is available through the connected attack methods.

Source

Research source for this risk, when available.