---
title: "VIB-Probe — paper by Zhang et al. (2026) — summary and ideas"
description: "Vision-Language Models often describe things that are not actually in the image. 52 references, cited by 1 paper here."
authors: ["Feiran Zhang", "Yixin Wu", "Zhenghua Wang", "Xiaohua Wang", "Changze Lv", "Xuanjing Huang", "Xiaoqing Zheng"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.1078/"
doi: "10.18653/v1/2026.acl-long.1078"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/zhang-2026-vib-probe-detecting-mitigating-hallucinations/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhang-2026-vib-probe-detecting-mitigating-hallucinations/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/6/) / VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck

# VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck

**Authors:** [Feiran Zhang](https://md.research.elamur.ai/free/en/authors/feiran-zhang/), [Yixin Wu](https://md.research.elamur.ai/free/en/authors/yixin-wu/), [Zhenghua Wang](https://md.research.elamur.ai/free/en/authors/zhenghua-wang/), [Xiaohua Wang](https://md.research.elamur.ai/free/en/authors/xiaohua-wang/), [Changze Lv](https://md.research.elamur.ai/free/en/authors/changze-lv/), [Xuanjing Huang](https://md.research.elamur.ai/free/en/authors/xuanjing-huang/) and [Xiaoqing Zheng](https://md.research.elamur.ai/free/en/authors/xiaoqing-zheng/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.1078/) · [PDF](https://aclanthology.org/2026.acl-long.1078.pdf) · [DOI 10.18653/v1/2026.acl-long.1078](https://doi.org/10.18653/v1/2026.acl-long.1078) · [OpenAlex](https://openalex.org/W7166865727)

**Citations:** 2 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Vision-Language Models often describe things that are not actually in the image. Most existing detectors look only at output confidence, such as logit entropy, or rely on external verifiers. VIB-Probe looks inside the model instead. During decoding, it collects the raw per-head attention outputs from every Transformer layer for each generated token. A probe built on the Variational Information Bottleneck then compresses this large tensor into a small latent code. A KL term pushes the code to drop syntactic and visual noise, and a BCE loss trains it to predict a hallucination label.

The authors also use the probe's gradients to identify which attention heads most strongly drive hallucinated outputs. At inference time, when predicted risk passes a threshold, these heads are scaled down and the token is regenerated. This mitigation step needs no retraining.

On POPE, AMBER, COCO and M-HalDetect, the method outperforms DHCP, RepProbing and MetaToken in detection. It is robust to image perturbations and to domain shift, and it also reduces CHAIR hallucination scores. The approach requires white-box access to the model.

## Key points
- **Detection setup.** Evaluation covers six benchmarks using AUPRC: POPE in Random, Popular and Adversarial splits, plus AMBER, COCO and M-HalDetect. Tested models include LLaVA-v1.5-7B, LLaVA-v1.6-7B and Qwen2.5-VL-7B-Instruct.
- **KL ablation.** Removing the KL constraint brings performance down to roughly the RepProbing baseline.
  - LLaVA-v1.5-7B: POPE AUPRC falls from 96.96 to 88.32, and M-HalDetect from 82.35 to 71.91.
  - Qwen2.5-VL-7B-Instruct: M-HalDetect falls from 80.79 to 67.34.
- **Layer choice.** Drawing features from all 32 layers works best. Using only layers 1–8 drops POPE AUPRC to 68.71, and deeper layers are more informative than shallow ones.
- **Mitigation on LLaVA-v1.5-7B.** On a 500-image COCO val2014 subset, CHAIR_s falls from 59.3 (vanilla) to 44.9. This beats both VCD and PAI.
- **Mitigation on LLaVA-v1.6-7B.** POPE F1 reaches 89.5 and CHAIR_i reaches 8.7.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Fudan University](https://md.research.elamur.ai/free/en/universities/fudan-university/) — University, Shanghai, China · 167 works · 20,987 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 657 works

## Cite

Cite as: Feiran Zhang, Yixin Wu, Zhenghua Wang, Xiaohua Wang, Changze Lv, Xuanjing Huang and Xiaoqing Zheng. 2026. VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{zhang2026vib,
  title = {{VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck}},
  author = {Feiran Zhang, Yixin Wu, Zhenghua Wang, Xiaohua Wang, Changze Lv, Xuanjing Huang and Xiaoqing Zheng},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1078},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1078}
}
```

- [← previous VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checking](https://md.research.elamur.ai/free/en/papers/rothermel-2026-veritas-first-dynamic-benchmark-multimodal/)
- [next → What Breaks Knowledge Graph based RAG? Benchmarking and Empirical Insights into Reasoning under Incomplete Knowledge](https://md.research.elamur.ai/free/en/papers/zhou-2026-what-breaks-knowledge-graph-based/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/6/)

## References

52 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Alexander A. Alemi, Ian Fischer, Joshua V. Dillon, and Kevin Murphy. 2017. Deep variational information bottleneck. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings. OpenReview.net. [find](https://scholar.google.com/scholar?q=Alexander+A%2E+Alemi%2C+Ian+Fischer%2C+Joshua+V%2E+Dillon%2C+and+Kevin+Murphy%2E+2017%2E+Deep+variational+information+bottleneck%2E+In+5th+International+Conference+on+Learning+Representations%2C+ICLR+2017%2C+Toulon%2C+Fran)
2. Lynton Ardizzone, Radek Mackowiak, Carsten Rother, and Ullrich Köthe. 2020. Training normalizing flows with the information bottleneck for competitive generative classification. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020,… [find](https://scholar.google.com/scholar?q=Lynton+Ardizzone%2C+Radek+Mackowiak%2C+Carsten+Rother%2C+and+Ullrich+K%C3%B6the%2E+2020%2E+Training+normalizing+flows+with+the+information+bottleneck+for+competitive+generative+classification%2E+In+Advances+in+Neural+)
3. Yung-Sung Chuang, Linlu Qiu, Cheng-Yu Hsieh, Ranjay Krishna, Yoon Kim, and James Glass. 2024. [Lookback lens: Detecting and mitigating contextual hallucinations in large language models using only attention maps](https://md.research.elamur.ai/free/en/papers/chuang-2024-lookback-lens-detecting-mitigating-contextual/). Preprint, arXiv:2407.07071.
4. Jesse Davis and Mark Goadrich. 2006. The relationship between precision-recall and roc curves. In Proceedings of the 23rd international conference on Machine learning - ICML ’06. [find](https://scholar.google.com/scholar?q=Jesse+Davis+and+Mark+Goadrich%2E+2006%2E+The+relationship+between+precision%2Drecall+and+roc+curves%2E+In+Proceedings+of+the+23rd+international+conference+on+Machine+learning+%2D+ICML+%E2%80%9906%2E)
5. Laura Fieback, Nishilkumar Balar, Jakob Spiegelberg, and Hanno Gottschalk. 2025a. Efficient contrastive decoding with probabilistic hallucination detection - mitigating hallucinations in large vision language models -. arXiv preprint. [find](https://scholar.google.com/scholar?q=Laura+Fieback%2C+Nishilkumar+Balar%2C+Jakob+Spiegelberg%2C+and+Hanno+Gottschalk%2E+2025a%2E+Efficient+contrastive+decoding+with+probabilistic+hallucination+detection+%2D+mitigating+hallucinations+in+large+vision+)
6. Laura Fieback, Jakob Spiegelberg, and Hanno Gottschalk. Metatoken: Detecting hallucination in image descriptions by meta classification. Proceedings Copyright, 126:137. [find](https://scholar.google.com/scholar?q=Laura+Fieback%2C+Jakob+Spiegelberg%2C+and+Hanno+Gottschalk%2E+Metatoken%3A+Detecting+hallucination+in+image+descriptions+by+meta+classification%2E+Proceedings+Copyright%2C+126%3A137%2E)
7. Laura Fieback, Jakob Spiegelberg, and Hanno Gottschalk. 2025b. Metatoken: Detecting hallucination in image descriptions by meta classification. In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, pages 126–137. SCITEPRE… [find](https://scholar.google.com/scholar?q=Laura+Fieback%2C+Jakob+Spiegelberg%2C+and+Hanno+Gottschalk%2E+2025b%2E+Metatoken%3A+Detecting+hallucination+in+image+descriptions+by+meta+classification%2E+In+Proceedings+of+the+20th+International+Joint+Conferenc)
8. Anisha Gunjal, Jihan Yin, and Erhan Bas. 2024. Detecting and preventing hallucinations in large vision language models. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 18135–18143. [find](https://scholar.google.com/scholar?q=Anisha+Gunjal%2C+Jihan+Yin%2C+and+Erhan+Bas%2E+2024%2E+Detecting+and+preventing+hallucinations+in+large+vision+language+models%2E+In+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+volume+38%2C+pag)
9. Jinghan He, Kuan Zhu, Haiyun Guo, Junfeng Fang, Zhenglin Hua, Yuheng Jia, Ming Tang, Tat-Seng Chua, and Jinqiao Wang. 2025. [Cracking the code of hallucination in lvlms with vision-aware head divergence](https://md.research.elamur.ai/free/en/papers/he-2025-cracking-code-hallucination-lvlms-vision/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volum…
10. Dan Hendrycks and Kevin Gimpel. 2017. A baseline for detecting misclassified and out-of-distribution examples in neural networks. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings. OpenReview.net. [find](https://scholar.google.com/scholar?q=Dan+Hendrycks+and+Kevin+Gimpel%2E+2017%2E+A+baseline+for+detecting+misclassified+and+out%2Dof%2Ddistribution+examples+in+neural+networks%2E+In+5th+International+Conference+on+Learning+Representations%2C+ICLR+2017)
11. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Ming-Hsuan Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, and 8 others. 2025. [Qwen2.5-vl technical report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen2-5-vl-technical/). CoRR, abs/2502.13923…
12. Qidong Huang, Xiaoyi Dong, Pan Zhang, Bin Wang, Conghui He, Jiaqi Wang, Dahua Lin, Weiming Zhang, and Nenghai Yu. 2024. Opera: Alleviating hallucination in multi-modal large language models via over-trust penalty and retrospection-allocation. In 2024 IEEE/CVF Conference on Computer Vision and Patter… [find](https://scholar.google.com/scholar?q=Qidong+Huang%2C+Xiaoyi+Dong%2C+Pan+Zhang%2C+Bin+Wang%2C+Conghui+He%2C+Jiaqi+Wang%2C+Dahua+Lin%2C+Weiming+Zhang%2C+and+Nenghai+Yu%2E+2024%2E+Opera%3A+Alleviating+hallucination+in+multi%2Dmodal+large+language+models+via+over%2Dt)
13. Keqin Chen, Zhao Zhang, Weili Zeng, Richong Zhang, Feng Zhu, and Rui Zhao. 2023. [Shikra: Unleashing multimodal llm’s referential dialogue magic](https://md.research.elamur.ai/free/en/papers/chen-2023-shikra-unleashing-multimodal-llm-s/). arXiv preprint.
14. Insu Jeon, Wonkwang Lee, Myeongjang Pyeon, and Gunhee Kim. 2025. IB-GAN: disentangled representation learning with information bottleneck generative adversarial networks. CoRR, abs/2510.20165. — [arXiv:2510.20165](https://arxiv.org/abs/2510.20165) [find](https://scholar.google.com/scholar?q=Insu+Jeon%2C+Wonkwang+Lee%2C+Myeongjang+Pyeon%2C+and+Gunhee+Kim%2E+2025%2E+IB%2DGAN%3A+disentangled+representation+learning+with+information+bottleneck+generative+adversarial+networks%2E+CoRR%2C+abs%2F2510%2E20165%2E+%E2%80%94+%5BarXi)
15. Nick Jiang, Anish Kachinthaya, Suzanne Petryk, and Yossi Gandelsman. 2025a. Interpreting and editing vision-language representations to mitigate hallucinations. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Nick+Jiang%2C+Anish+Kachinthaya%2C+Suzanne+Petryk%2C+and+Yossi+Gandelsman%2E+2025a%2E+Interpreting+and+editing+vision%2Dlanguage+representations+to+mitigate+hallucinations%2E+In+The+Thirteenth+International+Confere)
16. Qintong Li, Zhiyong Wu, Lingpeng Kong, and Wei Bi. 2023c. Explanation regeneration via information bottleneck. In Findings of the Association for Computational Linguistics: ACL 2023, Toronto, Canada, July 9-14, 2023, pages 12081–12102. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Qintong+Li%2C+Zhiyong+Wu%2C+Lingpeng+Kong%2C+and+Wei+Bi%2E+2023c%2E+Explanation+regeneration+via+information+bottleneck%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2023%2C+Toronto%2C+Canada%2C+)
17. Zhangqi Jiang, Junkai Chen, Beier Zhu, Tingjin Luo, Yankun Shen, and Xu Yang. 2025b. Devils in middle layers of large vision-language models: Interpreting, detecting and mitigating object hallucinations via attention lens. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2025,… [find](https://scholar.google.com/scholar?q=Zhangqi+Jiang%2C+Junkai+Chen%2C+Beier+Zhu%2C+Tingjin+Luo%2C+Yankun+Shen%2C+and+Xu+Yang%2E+2025b%2E+Devils+in+middle+layers+of+large+vision%2Dlanguage+models%3A+Interpreting%2C+detecting+and+mitigating+object+hallucinatio)
18. Xiang Lisa Li and Jason Eisner. 2019a. Specializing word embeddings (for parsing) by information bottleneck. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). Association… [find](https://scholar.google.com/scholar?q=Xiang+Lisa+Li+and+Jason+Eisner%2E+2019a%2E+Specializing+word+embeddings+%28for+parsing%29+by+information+bottleneck%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+Language+Processing+a)
19. Diederik P. Kingma and Max Welling. 2014. Autoencoding variational bayes. In 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, April 14-16, 2014, Conference Track Proceedings. [find](https://scholar.google.com/scholar?q=Diederik+P%2E+Kingma+and+Max+Welling%2E+2014%2E+Autoencoding+variational+bayes%2E+In+2nd+International+Conference+on+Learning+Representations%2C+ICLR+2014%2C+Banff%2C+AB%2C+Canada%2C+April+14%2D16%2C+2014%2C+Conference+Track)
20. Jusung Lee, Sungguk Cha, Younghyun Lee, and Che-oljong Yang. 2024. Visual question answering instruction: Unlocking multimodal large language model to domain-specific visual multitasks. CoRR, abs/2402.08360. — [arXiv:2402.08360](https://arxiv.org/abs/2402.08360) [find](https://scholar.google.com/scholar?q=Jusung+Lee%2C+Sungguk+Cha%2C+Younghyun+Lee%2C+and+Che%2Doljong+Yang%2E+2024%2E+Visual+question+answering+instruction%3A+Unlocking+multimodal+large+language+model+to+domain%2Dspecific+visual+multitasks%2E+CoRR%2C+abs%2F2402)
21. Sicong Leng, Hang Zhang, Guanzheng Chen, Xin Li, Shijian Lu, Chunyan Miao, and Lidong Bing. 2024a. [Mitigating object hallucinations in large vision-language models through visual contrastive decoding](https://md.research.elamur.ai/free/en/papers/leng-2023-mitigating-object-hallucinations-large-vision/). In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, WA, USA, Jun…
22. Sicong Leng, Hang Zhang, Guanzheng Chen, Xin Li, Shijian Lu, Chunyan Miao, and Lidong Bing. 2024b. [Mitigating object hallucinations in large vision-language models through visual contrastive decoding](https://md.research.elamur.ai/free/en/papers/leng-2023-mitigating-object-hallucinations-large-vision/). In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13872–1…
23. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023a. [Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models](https://md.research.elamur.ai/free/en/papers/li-2023-blip-2-bootstrapping-language-image/). In International conference on machine learning, pages 19730–19742. PMLR.
24. Junnan Li, Dongxu Li, Silvio Savarese, and Steven C. H. Hoi. 2023b. [BLIP-2: bootstrapping language-image pre-training with frozen image encoders and large language models](https://md.research.elamur.ai/free/en/papers/li-2023-blip-2-bootstrapping-language-image/). In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of…
25. Qing Li, Jiahui Geng, Chenyang Lyu, Derui Zhu, Maxim Panov, and Fakhri Karray. 2024. Reference-free hallucination detection for large vision-language models. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4542–4551. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Qing+Li%2C+Jiahui+Geng%2C+Chenyang+Lyu%2C+Derui+Zhu%2C+Maxim+Panov%2C+and+Fakhri+Karray%2E+2024%2E+Reference%2Dfree+hallucination+detection+for+large+vision%2Dlanguage+models%2E+In+Findings+of+the+Association+for+Computa)
26. Xiang Lisa Li and Jason Eisner. 2019b. Specializing word embeddings (for parsing) by information bottleneck. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). [find](https://scholar.google.com/scholar?q=Xiang+Lisa+Li+and+Jason+Eisner%2E+2019b%2E+Specializing+word+embeddings+%28for+parsing%29+by+information+bottleneck%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+Language+Processing+a)
27. Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Wayne Xin Zhao, and Ji-Rong Wen. 2023d. [Evaluating object hallucination in large vision-language models](https://md.research.elamur.ai/free/en/papers/li-2023-evaluating-object-hallucination-large-vision/). In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 292–305.
28. Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. 2014. [Microsoft coco: Common objects in context](https://md.research.elamur.ai/free/en/papers/lin-2014-microsoft-coco-common-objects-context/). In European conference on computer vision, pages 740–755. Springer.
29. Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. 2024a. [Improved baselines with visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-improved-baselines-visual-instruction-tuning/). In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 26296–26306.
30. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. [Visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/). CoRR, abs/2304.08485.
31. Shi Liu, Kecheng Zheng, and Wei Chen. 2024b. Paying more attention to image: A training-free method for alleviating hallucination in lvlms. In Computer Vision - ECCV 2024 - 18th European Conference, Milan, Italy, September 29-October 4, 2024, Proceedings, Part LXXXIII, volume 15141 of Lecture Notes… [find](https://scholar.google.com/scholar?q=Shi+Liu%2C+Kecheng+Zheng%2C+and+Wei+Chen%2E+2024b%2E+Paying+more+attention+to+image%3A+A+training%2Dfree+method+for+alleviating+hallucination+in+lvlms%2E+In+Computer+Vision+%2D+ECCV+2024+%2D+18th+European+Conference%2C+M)
32. Shi Liu, Kecheng Zheng, and Wei Chen. 2024c. Paying more attention to image: A training-free method for alleviating hallucination in lvlms. In European Conference on Computer Vision, pages 125–140. Springer. [find](https://scholar.google.com/scholar?q=Shi+Liu%2C+Kecheng+Zheng%2C+and+Wei+Chen%2E+2024c%2E+Paying+more+attention+to+image%3A+A+training%2Dfree+method+for+alleviating+hallucination+in+lvlms%2E+In+European+Conference+on+Computer+Vision%2C+pages+125%E2%80%93140%2E+Sp)
33. Rabeeh Karimi Mahabadi, Yonatan Belinkov, and James Henderson. 2021. Variational information bottleneck for effective low-resource fine-tuning. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net. [find](https://scholar.google.com/scholar?q=Rabeeh+Karimi+Mahabadi%2C+Yonatan+Belinkov%2C+and+James+Henderson%2E+2021%2E+Variational+information+bottleneck+for+effective+low%2Dresource+fine%2Dtuning%2E+In+9th+International+Conference+on+Learning+Representati)
34. Vishnu Prabhakaran, Purav Aggarwal, Vinay Kumar Verma, Gokul Swamy, and Anoop Saladi. 2025. VADE: Visual attention guided hallucination detection and elimination. In Findings of the Association for Computational Linguistics: ACL 2025, pages 14949–14965, Vienna, Austria. Association for Computational… [find](https://scholar.google.com/scholar?q=Vishnu+Prabhakaran%2C+Purav+Aggarwal%2C+Vinay+Kumar+Verma%2C+Gokul+Swamy%2C+and+Anoop+Saladi%2E+2025%2E+VADE%3A+Visual+attention+guided+hallucination+detection+and+elimination%2E+In+Findings+of+the+Association+for+Co)
35. Anna Rohrbach, Lisa Anne Hendricks, Kaylee Burns, Trevor Darrell, and Kate Saenko. 2018. [Object hallucination in image captioning](https://md.research.elamur.ai/free/en/papers/rohrbach-2018-object-hallucination-image-captioning/). In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.
36. Qingyun Sun, Jianxin Li, Hao Peng, Jia Wu, Xingcheng Fu, Cheng Ji, and Philip S. Yu. 2022. Graph structure learning with variational information bottleneck. In Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Fourth Conference on Innovative Applications of Artificial Intell… [find](https://scholar.google.com/scholar?q=Qingyun+Sun%2C+Jianxin+Li%2C+Hao+Peng%2C+Jia+Wu%2C+Xingcheng+Fu%2C+Cheng+Ji%2C+and+Philip+S%2E+Yu%2E+2022%2E+Graph+structure+learning+with+variational+information+bottleneck%2E+In+Thirty%2DSixth+AAAI+Conference+on+Artifici)
37. Feilong Tang, Chengzhi Liu, Zhongxing Xu, Ming Hu, Zile Huang, Haochen Xue, Ziyang Chen, Zelin Peng, Zhiwei Yang, Sijin Zhou, Wenxue Li, Yulong Li, Wenxuan Song, Shiyan Su, Wei Feng, Jionglong Su, Mingquan Lin, Yifan Peng, Xuelian Cheng, and 2 others. 2025. Seeing far and clearly: Mitigating halluci… [find](https://scholar.google.com/scholar?q=Feilong+Tang%2C+Chengzhi+Liu%2C+Zhongxing+Xu%2C+Ming+Hu%2C+Zile+Huang%2C+Haochen+Xue%2C+Ziyang+Chen%2C+Zelin+Peng%2C+Zhiwei+Yang%2C+Sijin+Zhou%2C+Wenxue+Li%2C+Yulong+Li%2C+Wenxuan+Song%2C+Shiyan+Su%2C+Wei+Feng%2C+Jionglong+Su%2C+Min)
38. Naftali Tishby, Fernando C. N. Pereira, and William Bialek. 2000. The information bottleneck method. CoRR, physics/0004057. [find](https://scholar.google.com/scholar?q=Naftali+Tishby%2C+Fernando+C%2E+N%2E+Pereira%2C+and+William+Bialek%2E+2000%2E+The+information+bottleneck+method%2E+CoRR%2C+physics%2F0004057%2E)
39. Junyang Wang, Yuhang Wang, Guohai Xu, Jing Zhang, Yukai Gu, Haitao Jia, Jiaqi Wang, Haiyang Xu, Ming Yan, Ji Zhang, and 1 others. 2023. Amber: An llm-free multi-dimensional benchmark for mllms hallucination evaluation. arXiv preprint arXiv:2311.07397. [find](https://scholar.google.com/scholar?q=Junyang+Wang%2C+Yuhang+Wang%2C+Guohai+Xu%2C+Jing+Zhang%2C+Yukai+Gu%2C+Haitao+Jia%2C+Jiaqi+Wang%2C+Haiyang+Xu%2C+Ming+Yan%2C+Ji+Zhang%2C+and+1+others%2E+2023%2E+Amber%3A+An+llm%2Dfree+multi%2Ddimensional+benchmark+for+mllms+halluci)
40. Tianyun Yang, Ziniu Li, Juan Cao, and Chang Xu. 2025. Understanding and mitigating hallucination in large vision-language models via modular attribution and intervention. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Tianyun+Yang%2C+Ziniu+Li%2C+Juan+Cao%2C+and+Chang+Xu%2E+2025%2E+Understanding+and+mitigating+hallucination+in+large+vision%2Dlanguage+models+via+modular+attribution+and+intervention%2E+In+The+Thirteenth+Internation)
41. Qinghao Ye, Haiyang Xu, Guohai Xu, Jiabo Ye, Ming Yan, Yiyang Zhou, Junyang Wang, Anwen Hu, Pengcheng Shi, Yaya Shi, Chenliang Li, Yuanhong Xu, Hehong Chen, Junfeng Tian, Qian Qi, Ji Zhang, and Fei Huang. 2023. [mplug-owl: Modularization empowers large language models with multimodality](https://md.research.elamur.ai/free/en/papers/ye-2023-mplug-owl-modularization-empowers-large/). CoRR, abs/23…
42. Shukang Yin, Chaoyou Fu, Sirui Zhao, Tong Xu, Hao Wang, Dianbo Sui, Yunhang Shen, Ke Li, Xing Sun, and Enhong Chen. 2024. Woodpecker: hallucination correction for multimodal large language models. Sci. China Inf. Sci., 67(12). [find](https://scholar.google.com/scholar?q=Shukang+Yin%2C+Chaoyou+Fu%2C+Sirui+Zhao%2C+Tong+Xu%2C+Hao+Wang%2C+Dianbo+Sui%2C+Yunhang+Shen%2C+Ke+Li%2C+Xing+Sun%2C+and+Enhong+Chen%2E+2024%2E+Woodpecker%3A+hallucination+correction+for+multimodal+large+language+models%2E+Sci)
43. Jinrui Zhang, Teng Wang, Haigang Zhang, Ping Lu, and Feng Zheng. 2024. Reflective instruction tuning: Mitigating hallucinations in large vision-language models. In Computer Vision - ECCV 2024 - 18th European Conference, Milan, Italy, September 29-October 4, 2024, Proceedings, Part LXVIII, volume 151… [find](https://scholar.google.com/scholar?q=Jinrui+Zhang%2C+Teng+Wang%2C+Haigang+Zhang%2C+Ping+Lu%2C+and+Feng+Zheng%2E+2024%2E+Reflective+instruction+tuning%3A+Mitigating+hallucinations+in+large+vision%2Dlanguage+models%2E+In+Computer+Vision+%2D+ECCV+2024+%2D+18th+E)
44. Xiaofeng Zhang, Yihao Quan, Chen Shen, Chaochen Gu, Xiaosong Yuan, Shaotian Yan, Jiawei Cao, Hao Cheng, Kaijie Wu, and Jieping Ye. 2025a. [Shallow focus, deep fixes: Enhancing shallow layers vision attention sinks to alleviate hallucination in lvlms](https://md.research.elamur.ai/free/en/papers/zhang-2025-shallow-focus-deep-fixes-enhancing/). In Proceedings of the 2025 Conference on Empirical…
45. Yudong Zhang, Ruobing Xie, Xingwu Sun, Yiqing Huang, Jiansheng Chen, Zhanhui Kang, Di Wang, and Yu Wang. 2025b. Dhcp: Detecting hallucinations by cross-modal attention pattern in large vision-language models. In Proceedings of the 33rd ACM International Conference on Multimedia, pages 3555– 3564. [find](https://scholar.google.com/scholar?q=Yudong+Zhang%2C+Ruobing+Xie%2C+Xingwu+Sun%2C+Yiqing+Huang%2C+Jiansheng+Chen%2C+Zhanhui+Kang%2C+Di+Wang%2C+and+Yu+Wang%2E+2025b%2E+Dhcp%3A+Detecting+hallucinations+by+cross%2Dmodal+attention+pattern+in+large+vision%2Dlanguage)
46. Ge Zheng, Jiaye Qian, Jiajin Tang, and Sibei Yang. 2025. Why lvlms are more prone to hallucinations in longer responses: The role of context. CoRR, abs/2510.20229. — [arXiv:2510.20229](https://arxiv.org/abs/2510.20229) [find](https://scholar.google.com/scholar?q=Ge+Zheng%2C+Jiaye+Qian%2C+Jiajin+Tang%2C+and+Sibei+Yang%2E+2025%2E+Why+lvlms+are+more+prone+to+hallucinations+in+longer+responses%3A+The+role+of+context%2E+CoRR%2C+abs%2F2510%2E20229%2E+%E2%80%94+%5BarXiv%3A2510%2E20229%5D%28https%3A%2F%2Farxiv%2Eo)
47. Yiyang Zhou, Chenhang Cui, Jaehong Yoon, Linjun Zhang, Zhun Deng, Chelsea Finn, Mohit Bansal, and Huaxiu Yao. 2024. Analyzing and mitigating object hallucination in large vision-language models. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11… [find](https://scholar.google.com/scholar?q=Yiyang+Zhou%2C+Chenhang+Cui%2C+Jaehong+Yoon%2C+Linjun+Zhang%2C+Zhun+Deng%2C+Chelsea+Finn%2C+Mohit+Bansal%2C+and+Huaxiu+Yao%2E+2024%2E+Analyzing+and+mitigating+object+hallucination+in+large+vision%2Dlanguage+models%2E+In+Th)
48. Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, and Mohamed Elhoseiny. [Minigpt-4: Enhancing vision-language understanding with advanced large language models](https://md.research.elamur.ai/free/en/papers/zhu-2023-minigpt-4-enhancing-vision-language/). In The Twelfth International Conference on Learning Representations.
49. Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, and Mohamed Elhoseiny. 2024a. [Minigpt-4: Enhancing vision-language understanding with advanced large language models](https://md.research.elamur.ai/free/en/papers/zhu-2023-minigpt-4-enhancing-vision-language/). In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net.
50. Kun Zhu, Xiaocheng Feng, Xiyuan Du, Yuxuan Gu, Weijiang Yu, Haotian Wang, Qianglong Chen, Zheng Chu, Jingchang Chen, and Bing Qin. 2024b. An information bottleneck perspective for effective noise filtering on retrieval-augmented generation. In Proceedings of the 62nd Annual Meeting of the Associatio… [find](https://scholar.google.com/scholar?q=Kun+Zhu%2C+Xiaocheng+Feng%2C+Xiyuan+Du%2C+Yuxuan+Gu%2C+Weijiang+Yu%2C+Haotian+Wang%2C+Qianglong+Chen%2C+Zheng+Chu%2C+Jingchang+Chen%2C+and+Bing+Qin%2E+2024b%2E+An+information+bottleneck+perspective+for+effective+noise+filt)
51. Lanyun Zhu, Deyi Ji, Tianrun Chen, Peng Xu, Jieping Ye, and Jun Liu. 2025. IBD: alleviating hallucinations in large vision-language models via image-biased decoding. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2025, Nashville, TN, USA, June 11-15, 2025… [find](https://scholar.google.com/scholar?q=Lanyun+Zhu%2C+Deyi+Ji%2C+Tianrun+Chen%2C+Peng+Xu%2C+Jieping+Ye%2C+and+Jun+Liu%2E+2025%2E+IBD%3A+alleviating+hallucinations+in+large+vision%2Dlanguage+models+via+image%2Dbiased+decoding%2E+In+IEEE%2FCVF+Conference+on+Computer)
52. Geigh Zollicoffer, Minh Vu, and Manish Bhattarai. 2025. Mtre: Multi-token reliability estimation for hallucination detection in vlms. arXiv preprint. [find](https://scholar.google.com/scholar?q=Geigh+Zollicoffer%2C+Minh+Vu%2C+and+Manish+Bhattarai%2E+2025%2E+Mtre%3A+Multi%2Dtoken+reliability+estimation+for+hallucination+detection+in+vlms%2E+arXiv+preprint%2E)

*Updated 31 October 2026*