VL-Calibration is a reinforcement learning method that separates visual and reasoning confidence in LVLMs via intrinsic visual certainty estimation to improve calibration and accuracy.
InThe Thirty-ninth Annual Conference on Neural Information Processing Systems
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
-
VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning
VL-Calibration is a reinforcement learning method that separates visual and reasoning confidence in LVLMs via intrinsic visual certainty estimation to improve calibration and accuracy.