Detecting and Preventing Hallucinations in Large Vision Language Models
Open accessAbstract
Instruction tuned Large Vision Language Models (LVLMs) have significantly advanced in generalizing across a diverse set of multi-modal tasks, especially for Visual Question Answering (VQA). However, generating detailed responses that are visually grounded is still a challenging task for these models. We find that even the current state-of-the-art LVLMs (InstructBLIP) still contain a staggering 30 percent of the hallucinatory text in the form of non-existent objects, unfaithful descriptions, and inaccurate relationships. To address this, we introduce M-HalDetect, a Multimodal Hallucination Detection Dataset that can be used to train and benchmark models for hallucination detection and prevention. M-HalDetect consists of 16k fine-grained annotations on VQA examples, making it the first comprehensive multi-modal hallucination detection dataset for detailed image descriptions. Unlike previous work that only consider object hallucination, we additionally annotate both entity descriptions and relationships that are unfaithful. To demonstrate the potential of this dataset for hallucination prevention, we optimize InstructBLIP through our novel Fine-grained Direct Preference Optimization (FDPO). We also train fine-grained multi-modal reward models from InstructBLIP and evaluate their effectiveness with best-of-n rejection sampling (RS). We perform human evaluation on both FDPO and rejection sampling, and find that they reduce hallucination rates in InstructBLIP by 41% and 55% respectively. We also find that our reward model generalizes to other multi-modal models, reducing hallucinations in LLaVA and mPLUG-OWL by 15% and 57% respectively, and has strong correlation with human evaluated accuracy scores. The dataset is available at https://github.com/hendryx-scale/mhal-detect.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Early work exists on detecting and preventing hallucinations in large vision-language models and other multimodal foundation models.
“There is also early work on detecting and preventing hallucinations in large vision language models [41] and other multimodal foundation models [128].”
From Grounding and Evaluation for Large Language Models: Practical Challenges and Lessons Learned (Survey) (2024) · cites this paper as [41]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460 · see citing works
Kaugeranna, R., Kaugeranna, E., & Aion, (. S. 4. 6. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.giscience.2023.43 · Cited by 13,695 · see citing works
Proximal Policy Optimization Algorithms
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017) · arXiv · doi:10.48550/arxiv.1707.06347 · Cited by 10,625 · see citing works
Survey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023) · ACM Computing Surveys · doi:10.1145/3571730 · Cited by 5,025 · see citing works
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208 · see citing works
InstructBLIP: Towards General-purpose Vision-Language Models with Instruction TuningPreprint
Dai, W., Li, J., Li, D., Tiong, A., Zhao, J.-Q., Wang, W., Li, B. A., Fung, P., & Hoi, S. (2023) · Neural Information Processing Systems · doi:10.48550/arxiv.2305.06500 · Cited by 4,191 · see citing works
Llama 2: Open Foundation and Fine-Tuned Chat Models
Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.09288 · Cited by 2,604 · see citing works
Evaluating Object Hallucination in Large Vision-Language ModelsPreprint
Li, Y., Du, Y., Zhou, K., et al. (2023) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2305.10355 · Cited by 2,148 · see citing works
Flamingo: a Visual Language Model for Few-Shot Learning
Alayrac, J.-B., Donahue, J., Luc, P., Miech, A., Barr, I., Hasson, Y., Lenc, K., Mensch, A., Millican, K., Reynolds, M., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.14198 · Cited by 1,555 · see citing works
mPLUG-Owl: Modularization Empowers Large Language Models with MultimodalityPreprint
Ye, Q.-H., Xu, H., Xu, G., Ye, J., Yan, M., Zhou, Y., Wang, J., Hu, A., Shi, P.-C., Shi, Y., et al. (2023) · arXiv.org · doi:10.48550/arxiv.2304.14178 · Cited by 1,318 · see citing works
Finetuned Language Models Are Zero-Shot Learners
Wei, J., Bosma, M., Zhao, V. Y., Guu, K., Yu, A. W., Lester, B., Du, N., Dai, A. M., & Le, Q. V. (2021) · arXiv · doi:10.48550/arxiv.2109.01652 · Cited by 1,101 · see citing works
Liu, H., Li, C., Wu, Q., & Lee, Y. J. (2023) · arXiv · doi:10.48550/arxiv.2304.08485 · Cited by 682 · see citing works
Evaluating Object Hallucination in Large Vision-Language Models
Li, Y., Du, Y., Zhou, K., Wang, J., Xin, Z., & Wen, J.-R. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.20 · Cited by 580 · see citing works
Learning to summarize from human feedback
Stiennon, N., Ouyang, L., Wu, J., Ziegler, D. M., Lowe, R., Voss, C., Radford, A., Amodei, D., & Christiano, P. (2020) · arXiv · doi:10.48550/arxiv.2009.01325 · Cited by 550 · see citing works
Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
Bai, Y., Jones, A., Ndousse, K., Askell, A., Chen, A., DasSarma, N., Drain, D., Fort, S., Ganguli, D., Henighan, T., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.05862 · Cited by 442 · see citing works
Fine-Tuning Language Models from Human Preferences
Ziegler, D. M., Stiennon, N., Wu, J., Brown, T. B., Radford, A., Amodei, D., Christiano, P., & Irving, G. (2019) · arXiv · doi:10.48550/arxiv.1909.08593 · Cited by 413 · see citing works
Bang, Y., Cahyawijaya, S., Lee, N., Dai, W., Su, D., Wilie, B., Lovenia, H., Ji, Z., Yu, T., Chung, W., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.04023 · Cited by 337 · see citing works
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C. D., & Finn, C. (2023) · arXiv · doi:10.48550/arxiv.2305.18290 · Cited by 293 · see citing works
WebGPT: Browser-assisted question-answering with human feedback
Nakano, R., Hilton, J., Balaji, S., Wu, J., Ouyang, L., Kim, C., Hesse, C., Jain, S., Kosaraju, V., Saunders, W., et al. (2021) · arXiv · doi:10.48550/arxiv.2112.09332 · Cited by 272 · see citing works
Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuning
Liu, F., Lin, K., Li, L., Wang, J., Yacoob, Y., & Wang, L. (2023) · arXiv · doi:10.48550/arxiv.2306.14565 · Cited by 149 · see citing works
Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO
Engstrom, L., Ilyas, A., Santurkar, S., Tsipras, D., Janoos, F., Rudolph, L., & Madry, A. (2020) · arXiv · doi:10.48550/arxiv.2005.12729 · Cited by 146 · see citing works
ROUGE 2.0: Updated and Improved Measures for Evaluation of Summarization TasksPreprint
Ganesan, K. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.01937 · Cited by 107 · see citing works
Lightman, H., Kosaraju, V., Burda, Y., Edwards, H., Baker, B., Lee, T., Leike, J., Schulman, J., Sutskever, I., & Cobbe, K. (2023) · arXiv · doi:10.48550/arxiv.2305.20050 · Cited by 65 · see citing works
Fine-Grained Human Feedback Gives Better Rewards for Language Model Training
Wu, Z., Hu, Y., Shi, W., Dziri, N., Suhr, A., Ammanabrolu, P., Smith, N. A., Ostendorf, M., & Hajishirzi, H. (2023) · arXiv · doi:10.48550/arxiv.2306.01693 · Cited by 40 · see citing works
Large Language Models are not Fair Evaluators
Wang, P., Li, L., Chen, L., Cai, Z., Zhu, D., Lin, B., Cao, Y., Liu, Q., Liu, T., & Sui, Z. (2023) · arXiv · doi:10.48550/arxiv.2305.17926 · Cited by 34 · see citing works
Large Multimodal Models: Notes on CVPR 2023 TutorialPreprint
Li, C. (2023) · arXiv.org · doi:10.48550/arxiv.2306.14895 · Cited by 24 · see citing works
Secrets of RLHF in Large Language Models Part I: PPOPreprint
Zheng, R., Dou, S., Gao, S., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.04964 · Cited by 19 · see citing works
OpenAI, Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.08774
Flagged first, then by citations
Research works printed but not linked: 3
- Bird, S.; Klein, E.; and Loper, E. 2009. Natural language processing with Python: analyzing text with the natural lan- guage toolkit. " O'Reilly Media, Inc.".
- Brown, T.; Mann, B.; Ryder, N.; Subbiah, M.; Kaplan, J. D.; Dhariwal, P.; Neelakantan, A.; Shyam, P.; Sastry, G.; Askell, A.; Agarwal, S.; Herbert-Voss, A.; Krueger, G.; Henighan, T.; Child, R.; Ramesh, A.; Ziegler, D.; Wu, J.; Winter, C.; Hesse, C.; Chen, M.; Sigler, E.; Litwin, M.; Gray, S.; Chess, B.; Clark, J.; Berner, C.; McCandlish, S.; Radford, A.; Sutskever, I.; and Amodei, D. 2020. Language Mod- els are Few-Shot Learners. Advances in Neural Information Processing Systems, 33: 1877-1901.
- Papineni, K.; Roukos, S.; Ward, T.; and Zhu, W.-J. 2002. Bleu: a Method for Automatic Evaluation of Machine Trans- lation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, 311-318. Asso- ciation for Computational Linguistics.
Cited by
Sorted by citations
Foundation Models Defining a New Era in Vision: A Survey and Outlook
Awais, M., Naseer, M., Khan, S., Anwer, R. M., Cholakkal, H., Shah, M., Yang, M.-H., & Khan, F. S. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3506283 · Cited by 304 · see citing works
Guan, T., Liu, F., Wu, X., Xian, R., Li, Z., Liu, X., Wang, X., Chen, L., Huang, F., Yacoob, Y., et al. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.01363 · Cited by 166 · see citing works
HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data
Yu, Q., Li, J., Wei, L., Pang, L., Ye, W., Qin, B., Tang, S., Tian, Q., & Zhuang, Y. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.01230 · Cited by 59 · see citing works
Kenthapadi, K., Sameki, M., & Taly, A. (2024) · Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · doi:10.1145/3637528.3671467 · Cited by 36 · see citing works
Exploring the Transferability of Visual Prompting for Multimodal Large Language Models
Zhang, Y., Dong, Y., Zhang, S., Min, T., Su, H., & Zhu, J. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.02508 · Cited by 17 · see citing works
Westhäußer, R., Minker, W., & Zepf, S. (2026) · Proceedings of the 31st International Conference on Intelligent User Interfaces · doi:10.1145/3742413.3789222 · Cited by 1 · see citing works
- Authors
- Gunjal, A., Yin, J., & Bas, E.
- ORCID
- Bas, Erhan
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2024-03-24
- Volume
- 38
- Issue
- 16
- Pages
- 18135-18143
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W4393160204
- Cited by
- 161
- References
- 28
- Citations by year
- 2026: 48 · 2025: 71 · 2024: 41 · 2023: 1
- FWCI
- 32.77
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- hallucination detection, hallucination prevention, large vision-language models, visual question answering, InstructBLIP, rejection sampling, LLaVA
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
