---
title: "Utility-Oriented Visual Evidence Selection for Multimodal"
description: "The authors instead define an image's value as the information gain it produces in the model's output distribution. 42 references."
authors: ["Weiqing Luo", "Zongye Hu", "Xiao Wang", "Zhiyuan Yu", "Haofeng Zhang", "Ziyi Huang"]
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.1620/"
doi: "10.18653/v1/2026.acl-long.1620"
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/luo-2026-utility-oriented-visual-evidence-selection/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/luo-2026-utility-oriented-visual-evidence-selection/)

[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/19/) / Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation

# Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation

**Authors:** Weiqing Luo, Zongye Hu, [Xiao Wang](https://md.research.elamur.ai/free/en/authors/xiao-wang/), [Zhiyuan Yu](https://md.research.elamur.ai/free/en/authors/zhiyuan-yu/), Haofeng Zhang and [Ziyi Huang](https://md.research.elamur.ai/free/en/authors/ziyi-huang/) · **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.1620/) · [PDF](https://aclanthology.org/2026.acl-long.1620.pdf) · [DOI 10.18653/v1/2026.acl-long.1620](https://doi.org/10.18653/v1/2026.acl-long.1620) · [OpenAlex](https://openalex.org/W7166815487)

**Citations:** 1 citations (OpenAlex, 2026-10-14)

## Summary
Multimodal RAG systems usually choose supporting images by semantic similarity, for example with CLIP-style encoders or MLLM relevance scorers. An image can look similar to the query and still not help answer it, such as a dog of the wrong breed that shares a striking pose. The authors instead define an image's value as the information gain it produces in the model's output distribution. Computing that gain over the full answer space is intractable and noisy, so they replace it with a binary latent "helpfulness" variable. They prove that, under mild assumptions, ranking candidates by information gain on this variable gives the same ordering as ranking in the answer space. For practical use, they add a training-free pipeline in which a small multimodal surrogate scores and ranks the candidate images. The large target model then runs only once to generate the answer. On MRAG-Bench and Visual-RAG, across several model families, the method beats state-of-the-art RAG baselines at lower computational cost. This makes utility-aware evidence selection practical.

## Key points
- Evidence utility is formalised as the KL divergence between the model's answer distribution with and without the evidence. Analysis is limited to candidates that do not harm generation, and the selection problem is reduced to a Bernoulli helpfulness estimate.
- Tested target/surrogate pairs include Qwen3-VL-8B with Qwen3-VL-2B, Ovis2.5-9B with Ovis2.5-2B, and Qwen2.5-VL-72B with 3B/7B surrogates.
- Surrogate false-positive rates are low. They range from 1.97% to 2.25% on MRAG-Bench (11,523 candidates) and from 2.22% to 3.50% on Visual-RAG (3,740 candidates).
- On a 100-sample check, small surrogates rank evidence much like the 72B model. On MRAG-Bench, Top-1 ground-truth hit rate is 80.49 for the 3B surrogate, 83.93 for 7B, and 83.96 for 72B. Visual-RAG shows a larger scale gap but similar trends.
- In most settings the method outperforms verbalized uncertainty quantification and listwise ranking, with gaps of up to about 12 points. Code is available at github.com/Hcnaeg/utility-mrag.

*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

- [Arizona State University](https://md.research.elamur.ai/free/en/universities/arizona-state-university/) — University, Tempe, United States · 50 works · 2,032 citations
- [Texas A&M University](https://md.research.elamur.ai/free/en/universities/texas-a-m-university/) — University, College Station, United States · 26 works · 1,871 citations

📚 Shelves (6): [🖥️ 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 · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,437 works

🧭 Topics (2): [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 657 works · [🔍 Information retrieval & search](https://md.research.elamur.ai/free/en/topics/information-retrieval/) · 473 works

## Cite

Cite as: Weiqing Luo, Zongye Hu, Xiao Wang, Zhiyuan Yu, Haofeng Zhang and Ziyi Huang. 2026. Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation. 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{luo2026utility,
  title = {{Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation}},
  author = {Weiqing Luo, Zongye Hu, Xiao Wang, Zhiyuan Yu, Haofeng Zhang and Ziyi Huang},
  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.1620},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1620}
}
```

- [← previous User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios](https://md.research.elamur.ai/free/en/papers/wu-2026-user-perceptions-vs-proxy-llm/)
- [next → UTokyo Tsuruoka Lab at SemEval-2026 Task 9: Efficient Single Forward Pass Inference for Multi-Label Polarization Classification](https://md.research.elamur.ai/free/en/papers/tangkulung-2026-utokyo-tsuruoka-lab-semeval-2026/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/19/)

## References

42 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, and 45 others. 2025. [Qwen3-vl technical report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen3-vl-technical-report/). Preprint, ar…
2. Boqi Chen, Anuj Khare, Gaurav Kumar, Arjun Akula, and Pradyumna Narayana. 2025. Seeing beyond: Enhancing visual question answering with multi-modal retrieval. In Proceedings of the 31st International Conference on Computational Linguistics: Industry Track, pages 410–421. [find](https://scholar.google.com/scholar?q=Boqi+Chen%2C+Anuj+Khare%2C+Gaurav+Kumar%2C+Arjun+Akula%2C+and+Pradyumna+Narayana%2E+2025%2E+Seeing+beyond%3A+Enhancing+visual+question+answering+with+multi%2Dmodal+retrieval%2E+In+Proceedings+of+the+31st+International+)
3. Wenhu Chen, Hexiang Hu, Xi Chen, Pat Verga, and William Cohen. 2022. [Murag: Multimodal retrieval-augmented generator for open question answering over images and text](https://md.research.elamur.ai/free/en/papers/chen-2022-murag-multimodal-retrieval-augmented-generator/). In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5558–5570.
4. Zhanpeng Chen, Chengjin Xu, Yiyan Qi, and Jian Guo. 2024. Mllm is a strong reranker: Advancing multimodal retrieval-augmented generation via knowledge-enhanced reranking and noise-injected training. arXiv preprint arXiv:2407.21439. — [arXiv:2407.21439](https://arxiv.org/abs/2407.21439) [find](https://scholar.google.com/scholar?q=Zhanpeng+Chen%2C+Chengjin+Xu%2C+Yiyan+Qi%2C+and+Jian+Guo%2E+2024%2E+Mllm+is+a+strong+reranker%3A+Advancing+multimodal+retrieval%2Daugmented+generation+via+knowledge%2Denhanced+reranking+and+noise%2Dinjected+training%2E+a)
5. Mehdi Cherti, Romain Beaumont, Ross Wightman, Mitchell Wortsman, Gabriel Ilharco, Cade Gordon, Christoph Schuhmann, Ludwig Schmidt, and Jenia Jitsev. 2023. [Reproducible scaling laws for contrastive language-image learning](https://md.research.elamur.ai/free/en/papers/cherti-2022-reproducible-scaling-laws-contrastive-language/). In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Reco…
6. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yixin Dai, Jiawei Sun, Haofen Wang, and Haofen Wang. 2023. [Retrieval-augmented generation for large language models: A survey](https://md.research.elamur.ai/free/en/papers/gao-2023-retrieval-augmented-generation-large-language/). arXiv preprint arXiv:2312.10997, 2(1).
7. Tiancheng Gu, Kaicheng Yang, Kaichen Zhang, Xiang An, Ziyong Feng, Yueyi Zhang, Weidong Cai, Jiankang Deng, and Lidong Bing. 2025. Unime-v2: Mllm-as-a-judge for universal multimodal embedding learning. arXiv preprint arXiv:2510.13515. — [arXiv:2510.13515](https://arxiv.org/abs/2510.13515) [find](https://scholar.google.com/scholar?q=Tiancheng+Gu%2C+Kaicheng+Yang%2C+Kaichen+Zhang%2C+Xiang+An%2C+Ziyong+Feng%2C+Yueyi+Zhang%2C+Weidong+Cai%2C+Jiankang+Deng%2C+and+Lidong+Bing%2E+2025%2E+Unime%2Dv2%3A+Mllm%2Das%2Da%2Djudge+for+universal+multimodal+embedding+learning)
8. Wenbo Hu, Jia-Chen Gu, Zi-Yi Dou, Mohsen Fayyaz, Pan Lu, Kai-Wei Chang, and Nanyun Peng. 2024. Mrag-bench: Vision-centric evaluation for retrieval-augmented multimodal models. arXiv preprint arXiv:2410.08182. — [arXiv:2410.08182](https://arxiv.org/abs/2410.08182) [find](https://scholar.google.com/scholar?q=Wenbo+Hu%2C+Jia%2DChen+Gu%2C+Zi%2DYi+Dou%2C+Mohsen+Fayyaz%2C+Pan+Lu%2C+Kai%2DWei+Chang%2C+and+Nanyun+Peng%2E+2024%2E+Mrag%2Dbench%3A+Vision%2Dcentric+evaluation+for+retrieval%2Daugmented+multimodal+models%2E+arXiv+preprint+arXiv%3A241)
9. Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig. 2021. [Scaling up visual and vision-language representation learning with noisy text supervision](https://md.research.elamur.ai/free/en/papers/jia-2021-scaling-up-visual-vision-language/). In International Conference on Machine Learning, pages 4904–4916. PMLR.
10. Ting Jiang, Minghui Song, Zihan Zhang, Haizhen Huang, Weiwei Deng, Feng Sun, Qi Zhang, Deqing Wang, and Fuzhen Zhuang. 2024a. E5-v: Universal embeddings with multimodal large language models. arXiv preprint arXiv:2407.12580. — [arXiv:2407.12580](https://arxiv.org/abs/2407.12580) [find](https://scholar.google.com/scholar?q=Ting+Jiang%2C+Minghui+Song%2C+Zihan+Zhang%2C+Haizhen+Huang%2C+Weiwei+Deng%2C+Feng+Sun%2C+Qi+Zhang%2C+Deqing+Wang%2C+and+Fuzhen+Zhuang%2E+2024a%2E+E5%2Dv%3A+Universal+embeddings+with+multimodal+large+language+models%2E+arXiv+pr)
11. Zhengbao Jiang, Frank F Xu, Luyu Gao, Zhiqing Sun, Qian Liu, Jane Dwivedi-Yu, Yiming Yang, Jamie Callan, and Graham Neubig. 2023. [Active retrieval augmented generation](https://md.research.elamur.ai/free/en/papers/jiang-2023-active-retrieval-augmented-generation/). In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7969–7992.
12. Ziyan Jiang, Rui Meng, Xinyi Yang, Semih Yavuz, Yingbo Zhou, and Wenhu Chen. 2024b. Vlm2vec: Training vision-language models for massive multimodal embedding tasks. arXiv preprint arXiv:2410.05160. — [arXiv:2410.05160](https://arxiv.org/abs/2410.05160) [find](https://scholar.google.com/scholar?q=Ziyan+Jiang%2C+Rui+Meng%2C+Xinyi+Yang%2C+Semih+Yavuz%2C+Yingbo+Zhou%2C+and+Wenhu+Chen%2E+2024b%2E+Vlm2vec%3A+Training+vision%2Dlanguage+models+for+massive+multimodal+embedding+tasks%2E+arXiv+preprint+arXiv%3A2410%2E05160%2E+%E2%80%94+)
13. Saurav Kadavath, Tom Conerly, Amanda Askell, Tom Henighan, Dawn Drain, Ethan Perez, Nicholas Schiefer, Zac Hatfield-Dodds, Nova DasSarma, Eli Tran-Johnson, and 1 others. 2022. Language models (mostly) know what they know. arXiv preprint arXiv:2207.05221. [find](https://scholar.google.com/scholar?q=Saurav+Kadavath%2C+Tom+Conerly%2C+Amanda+Askell%2C+Tom+Henighan%2C+Dawn+Drain%2C+Ethan+Perez%2C+Nicholas+Schiefer%2C+Zac+Hatfield%2DDodds%2C+Nova+DasSarma%2C+Eli+Tran%2DJohnson%2C+and+1+others%2E+2022%2E+Language+models+%28mostly%29)
14. Lorenz Kuhn, Yarin Gal, and Sebastian Farquhar. 2023. Semantic uncertainty: Linguistic invariances for uncertainty estimation in natural language generation. arXiv preprint arXiv:2302.09664. [find](https://scholar.google.com/scholar?q=Lorenz+Kuhn%2C+Yarin+Gal%2C+and+Sebastian+Farquhar%2E+2023%2E+Semantic+uncertainty%3A+Linguistic+invariances+for+uncertainty+estimation+in+natural+language+generation%2E+arXiv+preprint+arXiv%3A2302%2E09664%2E)
15. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, and 1 others. 2020. [Retrieval-augmented generation for knowledge-intensive NLP tasks](https://md.research.elamur.ai/free/en/papers/lewis-2020-retrieval-augmented-generation-knowledge-intensive/). Advances in Neural Information Processing Systems, 33:9459–…
16. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023. [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.
17. Junnan Li, Ramprasaath Selvaraju, Akhilesh Gotmare, Shafiq Joty, Caiming Xiong, and Steven Chu Hong Hoi. 2021. Align before fuse: Vision and language representation learning with momentum distillation. Advances in Neural Information Processing Systems, 34:9694–9705. [find](https://scholar.google.com/scholar?q=Junnan+Li%2C+Ramprasaath+Selvaraju%2C+Akhilesh+Gotmare%2C+Shafiq+Joty%2C+Caiming+Xiong%2C+and+Steven+Chu+Hong+Hoi%2E+2021%2E+Align+before+fuse%3A+Vision+and+language+representation+learning+with+momentum+distillation)
18. Sheng-Chieh Lin, Chankyu Lee, Mohammad Shoeybi, Jimmy Lin, Bryan Catanzaro, and Wei Ping. 2024. Mm-embed: Universal multimodal retrieval with multimodal llms. arXiv preprint arXiv:2411.02571. — [arXiv:2411.02571](https://arxiv.org/abs/2411.02571) [find](https://scholar.google.com/scholar?q=Sheng%2DChieh+Lin%2C+Chankyu+Lee%2C+Mohammad+Shoeybi%2C+Jimmy+Lin%2C+Bryan+Catanzaro%2C+and+Wei+Ping%2E+2024%2E+Mm%2Dembed%3A+Universal+multimodal+retrieval+with+multimodal+llms%2E+arXiv+preprint+arXiv%3A2411%2E02571%2E+%E2%80%94+%5BarXiv)
19. Zhen Lin, Shubhendu Trivedi, and Jimeng Sun. 2023. Generating with confidence: Uncertainty quantification for black-box large language models. arXiv preprint arXiv:2305.19187. [find](https://scholar.google.com/scholar?q=Zhen+Lin%2C+Shubhendu+Trivedi%2C+and+Jimeng+Sun%2E+2023%2E+Generating+with+confidence%3A+Uncertainty+quantification+for+black%2Dbox+large+language+models%2E+arXiv+preprint+arXiv%3A2305%2E19187%2E)
20. 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/). Advances in Neural Information Processing Systems, 36:34892– 34916.
21. Yikun Liu, Yajie Zhang, Jiayin Cai, Xiaolong Jiang, Yao Hu, Jiangchao Yao, Yanfeng Wang, and Weidi Xie. 2025. Lamra: Large multimodal model as your advanced retrieval assistant. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 4015–4025. [find](https://scholar.google.com/scholar?q=Yikun+Liu%2C+Yajie+Zhang%2C+Jiayin+Cai%2C+Xiaolong+Jiang%2C+Yao+Hu%2C+Jiangchao+Yao%2C+Yanfeng+Wang%2C+and+Weidi+Xie%2E+2025%2E+Lamra%3A+Large+multimodal+model+as+your+advanced+retrieval+assistant%2E+In+Proceedings+of+the+)
22. Shiyin Lu, Yang Li, Yu Xia, Yuwei Hu, Shanshan Zhao, Yanqing Ma, Zhichao Wei, Yinglun Li, Lunhao Duan, Jianshan Zhao, and 1 others. 2025. Ovis2. 5 technical report. arXiv preprint arXiv:2508.11737. — [arXiv:2508.11737](https://arxiv.org/abs/2508.11737) [find](https://scholar.google.com/scholar?q=Shiyin+Lu%2C+Yang+Li%2C+Yu+Xia%2C+Yuwei+Hu%2C+Shanshan+Zhao%2C+Yanqing+Ma%2C+Zhichao+Wei%2C+Yinglun+Li%2C+Lunhao+Duan%2C+Jianshan+Zhao%2C+and+1+others%2E+2025%2E+Ovis2%2E+5+technical+report%2E+arXiv+preprint+arXiv%3A2508%2E11737%2E+%E2%80%94+)
23. Lang Mei, Siyu Mo, Zhihan Yang, and Chong Chen. 2025. A survey of multimodal retrieval-augmented generation. arXiv preprint arXiv:2504.08748. — [arXiv:2504.08748](https://arxiv.org/abs/2504.08748) [find](https://scholar.google.com/scholar?q=Lang+Mei%2C+Siyu+Mo%2C+Zhihan+Yang%2C+and+Chong+Chen%2E+2025%2E+A+survey+of+multimodal+retrieval%2Daugmented+generation%2E+arXiv+preprint+arXiv%3A2504%2E08748%2E+%E2%80%94+%5BarXiv%3A2504%2E08748%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2504%2E08748%29)
24. Rui Meng, Ziyan Jiang, Ye Liu, Mingyi Su, Xinyi Yang, Yuepeng Fu, Can Qin, Zeyuan Chen, Ran Xu, Caiming Xiong, and 1 others. 2025. Vlm2vec-v2: Advancing multimodal embedding for videos, images, and visual documents. arXiv preprint arXiv:2507.04590. — [arXiv:2507.04590](https://arxiv.org/abs/2507.04590) [find](https://scholar.google.com/scholar?q=Rui+Meng%2C+Ziyan+Jiang%2C+Ye+Liu%2C+Mingyi+Su%2C+Xinyi+Yang%2C+Yuepeng+Fu%2C+Can+Qin%2C+Zeyuan+Chen%2C+Ran+Xu%2C+Caiming+Xiong%2C+and+1+others%2E+2025%2E+Vlm2vec%2Dv2%3A+Advancing+multimodal+embedding+for+videos%2C+images%2C+and+vi)
25. Matin Mortaheb, Mohammad A Amir Khojastepour, Srimat T Chakradhar, and Sennur Ulukus. 2025. Rag-check: Evaluating multimodal retrieval augmented generation performance. arXiv preprint arXiv:2501.03995. — [arXiv:2501.03995](https://arxiv.org/abs/2501.03995) [find](https://scholar.google.com/scholar?q=Matin+Mortaheb%2C+Mohammad+A+Amir+Khojastepour%2C+Srimat+T+Chakradhar%2C+and+Sennur+Ulukus%2E+2025%2E+Rag%2Dcheck%3A+Evaluating+multimodal+retrieval+augmented+generation+performance%2E+arXiv+preprint+arXiv%3A2501%2E03995)
26. Ikhtiyor Nematov, Tarik Kalai, Elizaveta Kuzmenko, Gabriele Fugagnoli, Dimitris Sacharidis, Katja Hose, and Tomer Sagi. 2025. Source attribution in retrieval-augmented generation. arXiv preprint arXiv:2507.04480. — [arXiv:2507.04480](https://arxiv.org/abs/2507.04480) [find](https://scholar.google.com/scholar?q=Ikhtiyor+Nematov%2C+Tarik+Kalai%2C+Elizaveta+Kuzmenko%2C+Gabriele+Fugagnoli%2C+Dimitris+Sacharidis%2C+Katja+Hose%2C+and+Tomer+Sagi%2E+2025%2E+Source+attribution+in+retrieval%2Daugmented+generation%2E+arXiv+preprint+arXiv)
27. Shintaro Ozaki, Yuta Kato, Siyuan Feng, Masayo Tomita, Kazuki Hayashi, Wataru Hashimoto, Ryoma Obara, Masafumi Oyamada, Katsuhiko Hayashi, Hidetaka Kamigaito, and 1 others. 2025. Understanding the impact of confidence in retrieval augmented generation: A case study in the medical domain. In Proceedi… [find](https://scholar.google.com/scholar?q=Shintaro+Ozaki%2C+Yuta+Kato%2C+Siyuan+Feng%2C+Masayo+Tomita%2C+Kazuki+Hayashi%2C+Wataru+Hashimoto%2C+Ryoma+Obara%2C+Masafumi+Oyamada%2C+Katsuhiko+Hayashi%2C+Hidetaka+Kamigaito%2C+and+1+others%2E+2025%2E+Understanding+the+imp)
28. Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, and 1 others. 2021. [Learning transferable visual models from natural language supervision](https://md.research.elamur.ai/free/en/papers/radford-2021-learning-transferable-visual-models-natural/). In International Conference on Machine Learning, pages 8748–8…
29. Heydar Soudani, Evangelos Kanoulas, and Faegheh Hasibi. 2025. Why uncertainty estimation methods fall short in rag: An axiomatic analysis. arXiv preprint arXiv:2505.07459. [find](https://scholar.google.com/scholar?q=Heydar+Soudani%2C+Evangelos+Kanoulas%2C+and+Faegheh+Hasibi%2E+2025%2E+Why+uncertainty+estimation+methods+fall+short+in+rag%3A+An+axiomatic+analysis%2E+arXiv+preprint+arXiv%3A2505%2E07459%2E)
30. Alon Talmor, Ori Yoran, Amnon Catav, Dan Lahav, Yizhong Wang, Akari Asai, Gabriel Ilharco, Hannaneh Hajishirzi, and Jonathan Berant. 2021. Multimodalqa: Complex question answering over text, tables and images. arXiv preprint arXiv:2104.06039. — [arXiv:2104.06039](https://arxiv.org/abs/2104.06039) [find](https://scholar.google.com/scholar?q=Alon+Talmor%2C+Ori+Yoran%2C+Amnon+Catav%2C+Dan+Lahav%2C+Yizhong+Wang%2C+Akari+Asai%2C+Gabriel+Ilharco%2C+Hannaneh+Hajishirzi%2C+and+Jonathan+Berant%2E+2021%2E+Multimodalqa%3A+Complex+question+answering+over+text%2C+tables+an)
31. Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, and 1 others. 2025. Gemma 3 technical report. arXiv preprint arXiv:2503.19786. [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Aishwarya+Kamath%2C+Johan+Ferret%2C+Shreya+Pathak%2C+Nino+Vieillard%2C+Ramona+Merhej%2C+Sarah+Perrin%2C+Tatiana+Matejovicova%2C+Alexandre+Ram%C3%A9%2C+Morgane+Rivi%C3%A8re%2C+and+1+others%2E+2025%2E+Gemma+3+technical+rep)
32. Michael Tschannen, Alexey Gritsenko, Xiao Wang, Muhammad Ferjad Naeem, Ibrahim Alabdulmohsin, Nikhil Parthasarathy, Talfan Evans, Lucas Beyer, Ye Xia, Basil Mustafa, and 1 others. 2025. Siglip 2: Multilingual vision-language encoders with improved semantic understanding, localization, and dense feat… [find](https://scholar.google.com/scholar?q=Michael+Tschannen%2C+Alexey+Gritsenko%2C+Xiao+Wang%2C+Muhammad+Ferjad+Naeem%2C+Ibrahim+Alabdulmohsin%2C+Nikhil+Parthasarathy%2C+Talfan+Evans%2C+Lucas+Beyer%2C+Ye+Xia%2C+Basil+Mustafa%2C+and+1+others%2E+2025%2E+Siglip+2%3A+Mult)
33. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025. [Internvl3](https://md.research.elamur.ai/free/en/papers/wang-2025-internvl3-5-advancing-open/). 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265.
34. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2022. [Self-consistency improves chain of thought reasoning in language models](https://md.research.elamur.ai/free/en/papers/wang-2022-self-consistency-improves-chain-thought/). arXiv preprint arXiv:2203.11171.
35. Tsung-Han Wu, Giscard Biamby, Jerome Quenum, Ritwik Gupta, Joseph E Gonzalez, Trevor Darrell, and David M Chan. 2024. Visual haystacks: A vision-centric needle-in-a-haystack benchmark. arXiv preprint arXiv:2407.13766. [find](https://scholar.google.com/scholar?q=Tsung%2DHan+Wu%2C+Giscard+Biamby%2C+Jerome+Quenum%2C+Ritwik+Gupta%2C+Joseph+E+Gonzalez%2C+Trevor+Darrell%2C+and+David+M+Chan%2E+2024%2E+Visual+haystacks%3A+A+vision%2Dcentric+needle%2Din%2Da%2Dhaystack+benchmark%2E+arXiv+preprint+)
36. Yin Wu, Quanyu Long, Jing Li, Jianfei Yu, and Wenya Wang. 2025. Visual-rag: Benchmarking text-to-image retrieval augmented generation for visual knowledge intensive queries. arXiv preprint arXiv:2502.16636. — [arXiv:2502.16636](https://arxiv.org/abs/2502.16636) [find](https://scholar.google.com/scholar?q=Yin+Wu%2C+Quanyu+Long%2C+Jing+Li%2C+Jianfei+Yu%2C+and+Wenya+Wang%2E+2025%2E+Visual%2Drag%3A+Benchmarking+text%2Dto%2Dimage+retrieval+augmented+generation+for+visual+knowledge+intensive+queries%2E+arXiv+preprint+arXiv%3A2502%2E)
37. Shan Xu, Zhaokun Yan, Chengxiao Dai, and Fan Wu. 2025. Mega-rag: a retrieval-augmented generation framework with multi-evidence guided answer refinement for mitigating hallucinations of llms in public health. Frontiers in Public Health, 13:1635381. [find](https://scholar.google.com/scholar?q=Shan+Xu%2C+Zhaokun+Yan%2C+Chengxiao+Dai%2C+and+Fan+Wu%2E+2025%2E+Mega%2Drag%3A+a+retrieval%2Daugmented+generation+framework+with+multi%2Devidence+guided+answer+refinement+for+mitigating+hallucinations+of+llms+in+public)
38. Zhengyuan Yang, Zhe Gan, Jianfeng Wang, Xiaowei Hu, Yumao Lu, Zicheng Liu, and Lijuan Wang. 2022. An empirical study of gpt-3 for few-shot knowledge-based vqa. In Proceedings of the AAAI conference on artificial intelligence, volume 36, pages 3081–3089. [find](https://scholar.google.com/scholar?q=Zhengyuan+Yang%2C+Zhe+Gan%2C+Jianfeng+Wang%2C+Xiaowei+Hu%2C+Yumao+Lu%2C+Zicheng+Liu%2C+and+Lijuan+Wang%2E+2022%2E+An+empirical+study+of+gpt%2D3+for+few%2Dshot+knowledge%2Dbased+vqa%2E+In+Proceedings+of+the+AAAI+conference+on)
39. Xiaoju Ye. 2023. calflops: a flops and params calculate tool for neural networks in pytorch framework. [find](https://scholar.google.com/scholar?q=Xiaoju+Ye%2E+2023%2E+calflops%3A+a+flops+and+params+calculate+tool+for+neural+networks+in+pytorch+framework%2E)
40. Tianyu Yu, Zefan Wang, Chongyi Wang, Fuwei Huang, Wenshuo Ma, Zhihui He, Tianchi Cai, Weize Chen, Yuxiang Huang, Yuanqian Zhao, and 1 others. 2025. [Minicpm-v 4.5: Cooking efficient mllms via architecture, data, and training recipe](https://md.research.elamur.ai/free/en/papers/yu-2025-minicpm-v-4-5/). arXiv preprint arXiv:2509.18154.
41. Xin Zhang, Yanzhao Zhang, Wen Xie, Mingxin Li, Ziqi Dai, Dingkun Long, Pengjun Xie, Meishan Zhang, Wenjie Li, and Min Zhang. 2024. Gme: Improving universal multimodal retrieval by multimodal LLMs. arXiv preprint arXiv:2412.16855. — [arXiv:2412.16855](https://arxiv.org/abs/2412.16855) [find](https://scholar.google.com/scholar?q=Xin+Zhang%2C+Yanzhao+Zhang%2C+Wen+Xie%2C+Mingxin+Li%2C+Ziqi+Dai%2C+Dingkun+Long%2C+Pengjun+Xie%2C+Meishan+Zhang%2C+Wenjie+Li%2C+and+Min+Zhang%2E+2024%2E+Gme%3A+Improving+universal+multimodal+retrieval+by+multimodal+LLMs%2E+arX)
42. Junjie Zhou, Yongping Xiong, Zheng Liu, Ze Liu, Shitao Xiao, Yueze Wang, Bo Zhao, Chen Jason Zhang, and Defu Lian. 2025. [Megapairs: Massive data synthesis for universal multimodal retrieval](https://md.research.elamur.ai/free/en/papers/zhou-2025-megapairs-massive-data-synthesis-universal/). In Proceedings of the 63rd Annual Meeting of the Association

*Updated 31 October 2026*