← All parts of this equation

Equation 1 · Part 2 · How Benchmark Contamination Actually Works in Agentic Evaluation

Symbol s_capability

sobserved=scapability+Δleak+Δchecker+Δfingerprints_{\text{observed}} = s_{\text{capability}} + \Delta_{\text{leak}} + \Delta_{\text{checker}} + \Delta_{\text{fingerprint}}
scapabilitys_{\text{capability}}

What this part means

the quantity anyone citing the score actually wants: the rate at which the policy would solve genuinely novel instances of this task class.

Its job in the formula

scs_capability is one of the signed contributions combined to compute the quantity on the left.

Where the article explains it

scapabilitys_{\text{capability}} is the quantity anyone citing the score actually wants: the rate at which the policy would solve genuinely novel instances of this task class.

The passage around this formula

It is useful to write down what an observed pass rate on an agentic benchmark is actually made of, because the rest of this article is a tour of the terms: sobserved=scapability+Δleak+Δchecker+Δfingerprints_{\text{observed}} = s_{\text{capability}} + \Delta_{\text{leak}} + \Delta_{\text{checker}} + \Delta_{\text{fingerprint}}. scapabilitys_{\text{capability}} is the quantity anyone citing the score actually wants: the rate at which the policy would solve genuinely novel instances of this task class. Δleak\Delta_{\text{leak}} is inflation from the environment itself disclosing the solution — through its history, its documentation, or a…

Read this part in the article →

Learn the underlying idea

A subscript is a label attached below a symbol. It often selects a time step, component, category, or member of a sequence.

Open the illustrated subscripts: which member of a family? guide →

See this notation across published equations →

The article lists its research sources here.