Detecting Misbehaviors of Large Vision-Language Models by Evidential Uncertainty Quantification

arXiv:2602.05535 · cs.LG · Submitted 2026-02-05 · Read on arXiv

cs.LG

Submitted: 2026-02-05

Updated: 2026-10-21

Terminology

• Hallucination — 3× in this paper · explained in 4 episodes
• Uncertainty Quantification — 3× in this paper · explained in 9 episodes
• Epistemic Uncertainty — 2× in this paper · explained in 12 episodes
• Evidential Uncertainty Quantification — 2× in this paper
• Vision-language models — 2× in this paper · explained in 2 episodes

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks