引導式測驗 · 中等 等級
莫拉維克悖論測驗
使用莫拉維克悖論來區分人類熟悉度和工程難度,並評估感知、計劃、身體動作和完整結果。
問題 1 的 8
A task feels effortless to a person. What does Moravec’s paradox caution against assuming?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- A task feels effortless to a person. What does Moravec’s paradox caution against assuming?
- That the task must be simple to engineer in a machine
- That people can perform the task
- That engineering methods can improve
- That the task has an observable outcome
- A game program receives a correct symbolic board position. What extra challenge appears when a robot faces the physical board?
- The written rules automatically change
- It must perceive the board and execute actions in the world
- Symbolic planning becomes unnecessary
- The camera supplies guaranteed perfect observations
- How should Moravec’s evolutionary explanation be treated?
- As a precise brain-to-computer speed formula
- As proof no robot can learn movement
- As a perspective on the contrast, not a universal capability law
- As evidence all abstract reasoning is easy
- A robot succeeds once on an uncluttered table. What is needed to assess reliable use in a changing workspace?
- Only a more polished demonstration video
- A chatbot’s description of how the robot works
- An assumption that all objects behave alike
- Repeated tests with relevant variation and recorded failures
- Why separate recognition failures from grasping failures?
- To understand which stage failed while still measuring complete task success
- To avoid testing the whole task
- To count every recognized object as a successful delivery
- To eliminate the need for feedback
- A language model explains how to pick up a cup. What does that demonstrate about a robot’s execution?
- That the physical task is already complete
- It does not by itself establish successful physical execution
- That every cup is safe to grasp
- That the model has measured the scene
- Which claim goes beyond the historical observation of uneven machine capabilities?
- Perception can involve difficult engineering
- Input assumptions matter in comparisons
- All physical tasks will remain impossible for machines
- A narrow success does not establish every capability
- A team publishes only successful robot clips. Which missing information limits the evaluation?
- The brand of the camera strap
- The background music title
- The number of words in the publicity release
- The total attempted trials, failures and tested conditions