A sandbox-trained multimodal search agent with process-oriented rewards transfers zero-shot to real Google Search and outperforms prior methods on FVQA, InfoSeek, and MMSearch.
In: Proceedings of the 37th International Conference on Machine Learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ProMMSearchAgent: A Generalizable Multimodal Search Agent Trained with Process-Oriented Rewards
A sandbox-trained multimodal search agent with process-oriented rewards transfers zero-shot to real Google Search and outperforms prior methods on FVQA, InfoSeek, and MMSearch.