CogAgent

Reported on 3 benchmarks across 3 tasks · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Visual Question Answering (VQA)onMM-Vet
GPT-4 score· 2023-12-14
52.8
best: 74.24 (MMCTAgent (GPT-4 + GPT-4V))
CogAgent: A Visual Language Model for GUI Agents arXiv:2312.08914
Visual Question AnsweringonMM-Vet
GPT-4 score· 2023-12-14
52.8
best: 74.24 (MMCTAgent (GPT-4 + GPT-4V))
CogAgent: A Visual Language Model for GUI Agents arXiv:2312.08914

Reasoning1 result

Natural Language Visual GroundingonScreenSpot
Accuracy (%)· 2023-12-14
47.4
best: 86.34 (UGround-V1-7B)
SOTA
CogAgent: A Visual Language Model for GUI Agents arXiv:2312.08914