A two-stage crop-and-predict framework improves high-resolution MLLM performance by using the model's own coarse localization to focus on a candidate region before final prediction.
Internvl: Scaling up vision foundation mod- els and aligning for generic visual-linguistic tasks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
A Training-Free, Task-Agnostic Framework for Enhancing MLLM Performance on High-Resolution Images
A two-stage crop-and-predict framework improves high-resolution MLLM performance by using the model's own coarse localization to focus on a candidate region before final prediction.