full transcript

From the Ted Talk by Fei-Fei Li: With spatial intelligence, AI will understand the real world

Unscramble the Blue Letters


A decade ago, ineaegmt from my lab enabled a database of millions of high-quality photos to help train computers to see. Today, we're doing the same with behaviors and actions to train computers and rbotos how to act in the 3D world. But instead of collecting static images, we develop simulation environments powered by 3D spatial mleods so that the computers can have iifnitne vaeiietrs of peisbilioitss to lraen to act. And you're just seeing a small nbemur of examples to taech our robots in a project led by my lab called Behavior.

We’re also making exciting progress in robotic language intelligence. Using large language model-based input, my students and our collaborators are among the first teams that can show a robotic arm performing a variety of tasks based on verbal instructions, like opening this drawer or unplugging a charged phone. Or making sandwiches, using beard, lettuce, tomatoes and even putting a nkiapn for the user. Typically I would like a little more for my snacdiwh, but this is a good start.

Open Cloze


A decade ago, ________ from my lab enabled a database of millions of high-quality photos to help train computers to see. Today, we're doing the same with behaviors and actions to train computers and ______ how to act in the 3D world. But instead of collecting static images, we develop simulation environments powered by 3D spatial ______ so that the computers can have ________ _________ of _____________ to _____ to act. And you're just seeing a small ______ of examples to _____ our robots in a project led by my lab called Behavior.

We’re also making exciting progress in robotic language intelligence. Using large language model-based input, my students and our collaborators are among the first teams that can show a robotic arm performing a variety of tasks based on verbal instructions, like opening this drawer or unplugging a charged phone. Or making sandwiches, using _____, lettuce, tomatoes and even putting a ______ for the user. Typically I would like a little more for my ________, but this is a good start.

Solution


  1. learn
  2. robots
  3. varieties
  4. teach
  5. infinite
  6. bread
  7. possibilities
  8. napkin
  9. models
  10. imagenet
  11. number
  12. sandwich

Original Text


A decade ago, ImageNet from my lab enabled a database of millions of high-quality photos to help train computers to see. Today, we're doing the same with behaviors and actions to train computers and robots how to act in the 3D world. But instead of collecting static images, we develop simulation environments powered by 3D spatial models so that the computers can have infinite varieties of possibilities to learn to act. And you're just seeing a small number of examples to teach our robots in a project led by my lab called Behavior.

We’re also making exciting progress in robotic language intelligence. Using large language model-based input, my students and our collaborators are among the first teams that can show a robotic arm performing a variety of tasks based on verbal instructions, like opening this drawer or unplugging a charged phone. Or making sandwiches, using bread, lettuce, tomatoes and even putting a napkin for the user. Typically I would like a little more for my sandwich, but this is a good start.

Frequently Occurring Word Combinations


ngrams of length 2

collocation frequency
spatial intelligence 6
virtuous cycle 2
making exciting 2
exciting progress 2
train computers 2
robotic arm 2
cambrian explosion 2

ngrams of length 3

collocation frequency
making exciting progress 2

Important Words


  1. act
  2. actions
  3. arm
  4. based
  5. behavior
  6. behaviors
  7. bread
  8. called
  9. charged
  10. collaborators
  11. collecting
  12. computers
  13. database
  14. decade
  15. develop
  16. drawer
  17. enabled
  18. environments
  19. examples
  20. exciting
  21. good
  22. imagenet
  23. images
  24. infinite
  25. input
  26. instructions
  27. intelligence
  28. lab
  29. language
  30. large
  31. learn
  32. led
  33. lettuce
  34. making
  35. millions
  36. models
  37. napkin
  38. number
  39. opening
  40. performing
  41. phone
  42. photos
  43. possibilities
  44. powered
  45. progress
  46. project
  47. putting
  48. robotic
  49. robots
  50. sandwich
  51. sandwiches
  52. show
  53. simulation
  54. small
  55. spatial
  56. start
  57. static
  58. students
  59. tasks
  60. teach
  61. teams
  62. today
  63. tomatoes
  64. train
  65. typically
  66. unplugging
  67. user
  68. varieties
  69. variety
  70. verbal
  71. world