AI agents on OSWorld take 2.7-4.3 times more steps than human trajectories, with latency rising sharply due to repeated large model calls for planning and reflection.
cc/paper_files/paper/2017/file/ 3f5ee243547dee91fbd053c1c4a845aa-Paper
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents
AI agents on OSWorld take 2.7-4.3 times more steps than human trajectories, with latency rising sharply due to repeated large model calls for planning and reflection.