List your company

ResearchFormerly WizWand

DriveLM

Benchmarks

TaskDatasetSOTA resultTrendResults

Autonomous driving reasoning (cross-view risk object perception, action prediction, and planning)

DriveLM

81Accuracy

25

Autonomous Driving Reasoning

DriveLM-ADvLM

49.89Final Score

24

Driving with language reasoning

DriveLM (leaderboard)

83Accuracy

12

DriveLM (test)

81Accuracy

11

Autonomous Driving Evaluation

DriveLM VRU-Accident benchmark

64.13GAR

10

Visual Question Answering

DriveLM

54.56BLEU-4

8

Graph Visual Question Answering

DriveLM GVQA

74Accuracy

7

Vision-Language Autonomous Driving

DriveLM open-loop

73.81Accuracy

6

Autonomous Driving

DriveLM

30Description Score

6

DriveLM (test)

59.01BLEU-4

6

Language Understanding

DriveLM

53.09BLEU-4

6

Driving Question Answering

DriveLM

81.23Accuracy

5

Autonomous Driving Reasoning

DriveLM 25% Token Retention

81.23Accuracy

5

Reasoning and Generation

DriveLM (test)

81Accuracy

5

Driving Scene Understanding

DriveLM (test)

45.45SPICE

5

Generative Question Answering

DriveLM (test)

53.09BLEU-4

5

DriveLM

67.3GPT Score

5

Drive Scene Understanding

DriveLM

0.9Params (B)

3

Spatial Perception

DriveLM

13.43Match Score

3