Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration

Large vision-language models (LVLMs) achieve impressive performance on multimodal tasks but often suffer from hallucination, and confidently describe objects or attributes not present in the image. Current inference-time interventions, while training-free, struggle to maintain accuracy in open-ended and long-form generation scenarios. We introduce the Confidence-Aware Attention Calibration (CAAC) framework to address this challenge by targeting two key biases: spatial perception bias, which distributes attention disproportionately across image tokens, and modality bias, which shifts focus from visual to textual inputs over time. CAAC employs a two-step approach: Visual-Token Calibration (VTC) to balance attention across visual tokens, and Adaptive Attention Re-Scaling (AAR) to reinforce visual grounding based on the model's confidence. This confidence-driven adjustment ensures consistent visual alignment during generation. Experiments on CHAIR, AMBER, and POPE benchmarks demonstrate that CAAC outperforms baselines, particularly in long-form generations, effectively reducing hallucination.

@article{fazli2025_2505.21472,
  title={ Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration },
  author={ Mehrdad Fazli and Bowen Wei and Ziwei Zhu },
  journal={arXiv preprint arXiv:2505.21472},
  year={ 2025 }
}