While Large Vision-Language Models (LVLMs) achieve remarkable success, hallucinations remain a significant barrier to their reliable deployment. Recent studies primarily attribute these issues to cros...
arXiv:2609.37263v1 Announce Type: new
Abstract: While Large Vision-Language Models (LVLMs) achieve remarkable success, hallucinations remain a significant barrier to their reliable deployment. Recent...
By Siqi Lu, Suo Wei, Yongbin Zheng, Jianhang Yao, Wanying Xu, Peng Wang
arXiv:2608. 07302v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) often suffer from object hallucination, generating objects that are absent from the image.
By Zichuan Wang, Songlin Yang, Bo Peng, Zhenchen Tang, Yang Li, Beibei Dong, Jing Dong
ReWEIGH the Evidence is a training‑free decoding technique that calibrates token‑level ordinal visual evidence to reduce hallucinations in large vision‑language models. It aggregates vocabulary ranks across visual positions, compares candidates to a token‑specific reference derived from unlabeled images, and applies a bounded penalty only when evidence falls below this reference. Experiments on four 7B backbones show up to a 21.3% reduction in hallucinated object mentions while largely preserving or improving descriptive and general performance, with minimal added latency.
By Jihae Jeong, Junha Choi, Hwanjo Yu
The paper introduces the concept of multi-view hallucination (MVH), where large vision-language models produce incorrect answers when processing images from multiple viewpoints. It presents MVH-Bench, a benchmark of 4.8k question-answer pairs that target cross-instance and cross-view hallucinations, and shows that MVH is common across recent models. The authors propose Reference Shift Contrastive Decoding (RSCD), a training-free decoding method that mitigates visual interference, achieving significant performance gains on MVH-Bench with LLaVA-OneVision and Qwen2.5-VL.
By Wooje Park, Insu Lee, Soohyun Kim, Jaeyun Jang, Minyoung Noh, Kyuhong Shim, Byonghyo Shim
arXiv:2608.29092v1 Announce Type: new
Abstract: Large vision-language models (LVLMs) frequently generate content unsupported by visual inputs. Preliminary experiments show that visual evidence is pri...
By Sihang Jia, Shuliang Liu, Songbo Yang, Xuming Hu