Reading Comprehension, Critical Inference & Para-Jumbles

Passage [Original Practice Problem · Modeled on GATE Pattern]:

'Autonomous driving systems rely on deep neural networks trained on millions of real-world driving hours. However, simulations can generate edge-case scenarios—such as extreme blizzard conditions with erratic pedestrian behavior—that occur only once every ten million highway miles. Therefore, autonomous vehicles trained with extensive simulation data will exhibit significantly lower accident rates in extreme weather than those trained on real-world driving alone.'

Question: Which of the following is an underlying assumption of the argument?

(A) Real-world driving data will soon become completely obsolete.

(B) Driving skills learned in simulated edge cases successfully transfer to physical vehicle control in real extreme weather.

(C) Blizzard conditions are the single most dangerous weather pattern.

(D) Neural networks train faster in simulated environments than real road miles.

Step 1: Identify conclusion and premise:

Premise: Simulations generate rare edge cases much more frequently than real driving.

Conclusion: Vehicles trained on simulations will have lower accident rates in physical extreme weather.

Step 2: Apply Assumption Negation Technique:

Negate option (B): 'Driving skills learned in simulated edge cases DO NOT transfer to physical vehicle control in real extreme weather.'

If simulated skills do not transfer to physical cars, simulation training cannot lower real accident rates.

The conclusion collapses completely.

Conclusion: (B) is the mandatory underlying assumption.