---
title: "VL-Calibration — paper by Xiao et al. (2026) — summary and ideas"
description: "The authors therefore build a pseudo-signal from two measures: Visual grounding: the KL-divergence between outputs on the… 47 references."
authors: ["Wenyi Xiao", "Xinchi XU", "Leilei Gan"]
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.2074/"
doi: "10.18653/v1/2026.acl-long.2074"
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/xiao-2026-vl-calibration-decoupled-confidence-calibration/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/xiao-2026-vl-calibration-decoupled-confidence-calibration/)

[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/55/) / VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning

# VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning

**Authors:** [Wenyi Xiao](https://md.research.elamur.ai/free/en/authors/wenyi-xiao/), Xinchi XU and [Leilei Gan](https://md.research.elamur.ai/free/en/authors/leilei-gan/) · **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.2074/) · [PDF](https://aclanthology.org/2026.acl-long.2074.pdf) · [DOI 10.18653/v1/2026.acl-long.2074](https://doi.org/10.18653/v1/2026.acl-long.2074) · [OpenAlex](https://openalex.org/W7166848639)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Vision-language models often give wrong answers while sounding certain. Existing methods that teach models to state their confidence were built for text-only LLMs and produce one overall score. For multimodal reasoning this score is ambiguous: an error may come from misreading the image or from faulty logic applied to a correct reading. Language priors can also hide visual uncertainty. VL-Calibration addresses this with a reinforcement learning (GRPO) setup in which the model first describes the image and states a visual confidence, then reasons and states a separate reasoning confidence. The two scores are combined by a harmonic mean, which keeps the overall confidence close to the weaker of the two.

No labels exist for perception correctness. The authors therefore build a pseudo-signal from two measures:
- **Visual grounding:** the KL-divergence between outputs on the original and a perturbed image.
- **Internal certainty:** token entropy.

This signal also reweights token-level advantages, so that visually uncertain tokens are penalized more when an answer is wrong. Across thirteen benchmarks, calibration improves sharply and accuracy rises as well. This matters for trustworthy multimodal systems in high-stakes domains.

## Key points
- On Qwen3-VL models, Expected Calibration Error (ECE) drops from 0.421 to 0.098. Average accuracy improves by 2.3–3.0% over the strongest baselines, including RLCR.
- The reward combines answer correctness, a Brier-style term for the holistic confidence, and a Brier-style term aligning visual confidence with the intrinsic visual certainty score.
- Experiments cover Qwen3-VL (4B–30B) and InternVL3.5-4B, with gains on out-of-distribution benchmarks. Models of 70B+ are untested.
- In a manual check of 1,000 sampled responses, the model gave clearly lower visual confidence to visually incorrect answers than to correct ones.
- When images were removed from DynaMath, the method showed the largest confidence gap between answerable and unanswerable problems. Code: github.com/Mr-Loevan/VL-Calibration.

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

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

🧭 Topics (3): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 811 works

## Cite

Cite as: Wenyi Xiao, Xinchi XU and Leilei Gan. 2026. VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning. 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{xiao2026vl,
  title = {{VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning}},
  author = {Wenyi Xiao, Xinchi XU and Leilei Gan},
  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.2074},
  url = {https://doi.org/10.18653/v1/2026.acl-long.2074}
}
```

- [← previous ViX-Ray: A Vietnamese Chest X-Ray Dataset for Vision-Language Models](https://md.research.elamur.ai/free/en/papers/nguyen-2026-vix-ray-vietnamese-chest-x/)
- [next → VLCE: A Knowledge-Enhanced Framework for Image Description in Disaster Assessment](https://md.research.elamur.ai/free/en/papers/rahman-2026-vlce-knowledge-enhanced-framework-image/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/55/)

## References

47 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Lukas Aichberger, Kajetan Schweighofer, Mykyta Ielanskyi, and Sepp Hochreiter. 2025. Improving uncertainty estimation through semantically diverse language generation. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Lukas+Aichberger%2C+Kajetan+Schweighofer%2C+Mykyta+Ielanskyi%2C+and+Sepp+Hochreiter%2E+2025%2E+Improving+uncertainty+estimation+through+semantically+diverse+language+generation%2E+In+The+Thirteenth+International+)
2. 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/). arXiv prepri…
3. Mehul Damani, Isha Puri, Stewart Slocum, Idan Shenfeld, Leshem Choshen, Yoon Kim, and Jacob Andreas. 2026. Beyond binary rewards: Training LMs to reason about their uncertainty. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Mehul+Damani%2C+Isha+Puri%2C+Stewart+Slocum%2C+Idan+Shenfeld%2C+Leshem+Choshen%2C+Yoon+Kim%2C+and+Jacob+Andreas%2E+2026%2E+Beyond+binary+rewards%3A+Training+LMs+to+reason+about+their+uncertainty%2E+In+The+Fourteenth+Inte)
4. Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, and Yarin Gal. 2024. Detecting hallucinations in large language models using semantic entropy. Nature, 630(8017):625–630. [find](https://scholar.google.com/scholar?q=Sebastian+Farquhar%2C+Jannik+Kossen%2C+Lorenz+Kuhn%2C+and+Yarin+Gal%2E+2024%2E+Detecting+hallucinations+in+large+language+models+using+semantic+entropy%2E+Nature%2C+630%288017%29%3A625%E2%80%93630%2E)
5. W Brier Glenn and 1 others. 1950. Verification of forecasts expressed in terms of probability. Monthly weather review, 78(1):1–3. [find](https://scholar.google.com/scholar?q=W+Brier+Glenn+and+1+others%2E+1950%2E+Verification+of+forecasts+expressed+in+terms+of+probability%2E+Monthly+weather+review%2C+78%281%29%3A1%E2%80%933%2E)
6. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2501.12948.
7. Yinghao Hu, Leilei Gan, Wenyi Xiao, Kun Kuang, and Fei Wu. 2025. Fine-tuning large language models for improving factuality in legal question answering. In Proceedings of the 31st international conference on computational linguistics, pages 4410–4427. [find](https://scholar.google.com/scholar?q=Yinghao+Hu%2C+Leilei+Gan%2C+Wenyi+Xiao%2C+Kun+Kuang%2C+and+Fei+Wu%2E+2025%2E+Fine%2Dtuning+large+language+models+for+improving+factuality+in+legal+question+answering%2E+In+Proceedings+of+the+31st+international+confer)
8. Chenchen Jing, Yuwei Wu, Xiaoxun Zhang, Yunde Jia, and Qi Wu. 2020. Overcoming language priors in vqa via decomposed linguistic representations. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 11181–11188. [find](https://scholar.google.com/scholar?q=Chenchen+Jing%2C+Yuwei+Wu%2C+Xiaoxun+Zhang%2C+Yunde+Jia%2C+and+Qi+Wu%2E+2020%2E+Overcoming+language+priors+in+vqa+via+decomposed+linguistic+representations%2E+In+Proceedings+of+the+AAAI+conference+on+artificial+int)
9. Gonzalez, Hao Zhang, and Ion Stoica. 2023. [Efficient memory management for large language model serving with pagedattention](https://md.research.elamur.ai/free/en/papers/kwon-2023-efficient-memory-management-large-language/). In Proceedings of the ACM SIGOPS 29th Symposium on Operating Systems Principles.
10. Jixuan Leng, Chengsong Huang, Banghua Zhu, and Jiaxin Huang. 2025. Taming overconfidence in llms: Reward calibration in rlhf. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Jixuan+Leng%2C+Chengsong+Huang%2C+Banghua+Zhu%2C+and+Jiaxin+Huang%2E+2025%2E+Taming+overconfidence+in+llms%3A+Reward+calibration+in+rlhf%2E+In+The+Thirteenth+International+Conference+on+Learning+Representations%2E)
11. Yibo Li, Miao Xiong, Jiaying Wu, and Bryan Hooi. 2025. Conftuner: Training large language models to express their confidence verbally. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Yibo+Li%2C+Miao+Xiong%2C+Jiaying+Wu%2C+and+Bryan+Hooi%2E+2025%2E+Conftuner%3A+Training+large+language+models+to+express+their+confidence+verbally%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Information+Proce)
12. Zhuowan Li, Xingrui Wang, Elias Stengel-Eskin, Adam Kortylewski, Wufei Ma, Benjamin Van Durme, and Alan L Yuille. 2023. Super-clevr: A virtual benchmark to diagnose domain robustness in visual reasoning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1496… [find](https://scholar.google.com/scholar?q=Zhuowan+Li%2C+Xingrui+Wang%2C+Elias+Stengel%2DEskin%2C+Adam+Kortylewski%2C+Wufei+Ma%2C+Benjamin+Van+Durme%2C+and+Alan+L+Yuille%2E+2023%2E+Super%2Dclevr%3A+A+virtual+benchmark+to+diagnose+domain+robustness+in+visual+reasoni)
13. Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, and Jianfeng Gao. 2024. [Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts](https://md.research.elamur.ai/free/en/papers/lu-2023-mathvista-evaluating-mathematical-reasoning-foundation/). In International Conference on Learning Representations.
14. Pan Lu, Ran Gong, Shibiao Jiang, Liang Qiu, Siyuan Huang, Xiaodan Liang, and Song-Chun Zhu. 2021. [Inter-gps: Interpretable geometry problem solving with formal language and symbolic reasoning](https://md.research.elamur.ai/free/en/papers/lu-2021-inter-gps-interpretable-geometry-problem/). In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th Int…
15. Fanqing Meng, Lingxiao Du, Zongkai Liu, Zhixiang Zhou, Quanfeng Lu, Daocheng Fu, Botian Shi, Wenhai Wang, Junjun He, Kaipeng Zhang, and 1 others. 2025. Mm-eureka: Exploring visual aha moment with rule-based large-scale reinforcement learning. arXiv preprint arXiv:2503.07365. [find](https://scholar.google.com/scholar?q=Fanqing+Meng%2C+Lingxiao+Du%2C+Zongkai+Liu%2C+Zhixiang+Zhou%2C+Quanfeng+Lu%2C+Daocheng+Fu%2C+Botian+Shi%2C+Wenhai+Wang%2C+Junjun+He%2C+Kaipeng+Zhang%2C+and+1+others%2E+2025%2E+Mm%2Deureka%3A+Exploring+visual+aha+moment+with+rule)
16. 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)
17. Runqi Qiao, Qiuna Tan, Guanting Dong, MinhuiWu MinhuiWu, Chong Sun, Xiaoshuai Song, Jiapeng Wang, Zhuoma Gongque, Shanglin Lei, Yifan Zhang, and 1 others. 2025. [We-math: Does your large multimodal model achieve human-like mathematical reasoning?](https://md.research.elamur.ai/free/en/papers/qiao-2024-math-large-multimodal-model-achieve/) In Proceedings of the 63rd Annual Meeting of the Assoc…
18. Zhewei Kang, Xuandong Zhao, and Dawn Song. 2025. Scalable best-of-n selection for large language models via self-certainty. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Zhewei+Kang%2C+Xuandong+Zhao%2C+and+Dawn+Song%2E+2025%2E+Scalable+best%2Dof%2Dn+selection+for+large+language+models+via+self%2Dcertainty%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Information+Processing+Syste)
19. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/). Advances in neural information processing systems, 36:53728–53741.
20. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph E. [find](https://scholar.google.com/scholar?q=Woosuk+Kwon%2C+Zhuohan+Li%2C+Siyuan+Zhuang%2C+Ying+Sheng%2C+Lianmin+Zheng%2C+Cody+Hao+Yu%2C+Joseph+E%2E)
21. Sainandan Ramakrishnan, Aishwarya Agrawal, and Stefan Lee. 2018. Overcoming language priors in visual question answering with adversarial regularization. Advances in neural information processing systems, 31. [find](https://scholar.google.com/scholar?q=Sainandan+Ramakrishnan%2C+Aishwarya+Agrawal%2C+and+Stefan+Lee%2E+2018%2E+Overcoming+language+priors+in+visual+question+answering+with+adversarial+regularization%2E+Advances+in+neural+information+processing+syst)
22. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347. [find](https://scholar.google.com/scholar?q=John+Schulman%2C+Filip+Wolski%2C+Prafulla+Dhariwal%2C+Alec+Radford%2C+and+Oleg+Klimov%2E+2017%2E+Proximal+policy+optimization+algorithms%2E+arXiv+preprint+arXiv%3A1707%2E06347%2E)
23. Dustin Schwenk, Apoorv Khandelwal, Christopher Clark, Kenneth Marino, and Roozbeh Mottaghi. 2022. [A-okvqa: A benchmark for visual question answering using world knowledge](https://md.research.elamur.ai/free/en/papers/schwenk-2022-okvqa-benchmark-visual-question-answering/). In European conference on computer vision, pages 146–162. Springer.
24. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 2024. [Deepseekmath: Pushing the limits of mathematical reasoning in open language models](https://md.research.elamur.ai/free/en/papers/shao-2024-deepseekmath-pushing-limits-mathematical-reasoning/). arXiv preprint arXiv:2402.03300.
25. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2024. [Hybridflow: A flexible and efficient rlhf framework](https://md.research.elamur.ai/free/en/papers/sheng-2024-hybridflow-flexible-efficient-rlhf-framework/). arXiv preprint arXiv: 2409.19256.
26. Fulin Shi, Wenyi Xiao, Bin Chen, Liang Din, and Leilei Gan. 2025. [Revealer: Reinforcement-guided visual reasoning for element-level text-image alignment evaluation](https://md.research.elamur.ai/free/en/papers/shi-2026-revealer-reinforcement-guided-visual-reasoning/). arXiv preprint arXiv:2512.23169.
27. Paul Stangel, David Bani-Harouni, Chantal Pellegrini, Ege Özsoy, Kamilia Zaripova, Matthias Keicher, and Nassir Navab. 2025. Rewarding doubt: A reinforcement learning approach to calibrated confidence expression of large language models. arXiv preprint arXiv:2503.02623. — [arXiv:2503.02623](https://arxiv.org/abs/2503.02623) [find](https://scholar.google.com/scholar?q=Paul+Stangel%2C+David+Bani%2DHarouni%2C+Chantal+Pellegrini%2C+Ege+%C3%96zsoy%2C+Kamilia+Zaripova%2C+Matthias+Keicher%2C+and+Nassir+Navab%2E+2025%2E+Rewarding+doubt%3A+A+reinforcement+learning+approach+to+calibrated+confidence)
28. Elias Stengel-Eskin, Peter Hase, and Mohit Bansal. 2024. Lacie: Listener-aware finetuning for calibration in large language models. Advances in Neural Information Processing Systems, 37:43080–43106. [find](https://scholar.google.com/scholar?q=Elias+Stengel%2DEskin%2C+Peter+Hase%2C+and+Mohit+Bansal%2E+2024%2E+Lacie%3A+Listener%2Daware+finetuning+for+calibration+in+large+language+models%2E+Advances+in+Neural+Information+Processing+Systems%2C+37%3A43080%E2%80%9343106%2E)
29. Roman Vashurin, Maiya Goloburda, Albina Ilina, Aleksandr Rubashevskii, Preslav Nakov, Artem Shelmanov, and Maxim Panov. 2025. Cocoa: A minimum bayes risk framework bridging confidence and consistency for uncertainty quantification in LLMs. In The Thirty-ninth Annual Conference on Neural Information… [find](https://scholar.google.com/scholar?q=Roman+Vashurin%2C+Maiya+Goloburda%2C+Albina+Ilina%2C+Aleksandr+Rubashevskii%2C+Preslav+Nakov%2C+Artem+Shelmanov%2C+and+Maxim+Panov%2E+2025%2E+Cocoa%3A+A+minimum+bayes+risk+framework+bridging+confidence+and+consistency+)
30. Haozhe Wang, Chao Qu, Zuming Huang, Wei Chu, Fangzhen Lin, and Wenhu Chen. 2025a. Vl-rethinker: Incentivizing self-reflection of vision-language models with reinforcement learning. arXiv preprint arXiv:2504.08837. [find](https://scholar.google.com/scholar?q=Haozhe+Wang%2C+Chao+Qu%2C+Zuming+Huang%2C+Wei+Chu%2C+Fangzhen+Lin%2C+and+Wenhu+Chen%2E+2025a%2E+Vl%2Drethinker%3A+Incentivizing+self%2Dreflection+of+vision%2Dlanguage+models+with+reinforcement+learning%2E+arXiv+preprint+arXi)
31. Ke Wang, Junting Pan, Weikang Shi, Zimu Lu, Houxing Ren, Aojun Zhou, Mingjie Zhan, and Hongsheng Li. 2024. [Measuring multimodal mathematical reasoning with math-vision dataset](https://md.research.elamur.ai/free/en/papers/wang-2024-measuring-multimodal-mathematical-reasoning-math/). In The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track.
32. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025b. [Internvl3.5: Advancing open-source multimodal models in versatility, reasoning, and efficiency](https://md.research.elamur.ai/free/en/papers/wang-2025-internvl3-5-advancing-open/). arXiv preprint arXiv:2508.18265.
33. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023. [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/). In The Eleventh International Conference on Learning Representations.
34. Wenyi Xiao and Leilei Gan. 2025. Fast-slow thinking GRPO for large vision-language model reasoning. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Wenyi+Xiao+and+Leilei+Gan%2E+2025%2E+Fast%2Dslow+thinking+GRPO+for+large+vision%2Dlanguage+model+reasoning%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Information+Processing+Systems%2E)
35. Wenyi Xiao, Ziwei Huang, Leilei Gan, Wanggui He, Haoyuan Li, Zhelun Yu, Fangxun Shu, Hao Jiang, and Linchao Zhu. 2025. Detecting and mitigating hallucination in large vision language models via fine-grained ai feedback. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pag… [find](https://scholar.google.com/scholar?q=Wenyi+Xiao%2C+Ziwei+Huang%2C+Leilei+Gan%2C+Wanggui+He%2C+Haoyuan+Li%2C+Zhelun+Yu%2C+Fangxun+Shu%2C+Hao+Jiang%2C+and+Linchao+Zhu%2E+2025%2E+Detecting+and+mitigating+hallucination+in+large+vision+language+models+via+fine%2Dg)
36. Wenyi Xiao, Zechuan Wang, Leilei Gan, Shuai Zhao, Zongrui Li, Ruirui Lei, Wanggui He, Luu Anh Tuan, Long Chen, Hao Jiang, and 1 others. 2024a. A comprehensive survey of direct preference optimization: Datasets, theories, variants, and applications. arXiv preprint arXiv:2410.15595. — [arXiv:2410.15595](https://arxiv.org/abs/2410.15595) [find](https://scholar.google.com/scholar?q=Wenyi+Xiao%2C+Zechuan+Wang%2C+Leilei+Gan%2C+Shuai+Zhao%2C+Zongrui+Li%2C+Ruirui+Lei%2C+Wanggui+He%2C+Luu+Anh+Tuan%2C+Long+Chen%2C+Hao+Jiang%2C+and+1+others%2E+2024a%2E+A+comprehensive+survey+of+direct+preference+optimization%3A)
37. Yijia Xiao, Edward Sun, Tianyu Liu, and Wei Wang. 2024b. Logicvista: Multimodal llm logical reasoning benchmark in visual contexts. arXiv preprint arXiv:2407.04973. — [arXiv:2407.04973](https://arxiv.org/abs/2407.04973) [find](https://scholar.google.com/scholar?q=Yijia+Xiao%2C+Edward+Sun%2C+Tianyu+Liu%2C+and+Wei+Wang%2E+2024b%2E+Logicvista%3A+Multimodal+llm+logical+reasoning+benchmark+in+visual+contexts%2E+arXiv+preprint+arXiv%3A2407%2E04973%2E+%E2%80%94+%5BarXiv%3A2407%2E04973%5D%28https%3A%2F%2Farxiv%2E)
38. Miao Xiong, Zhiyuan Hu, Xinyang Lu, YIFEI LI, Jie Fu, Junxian He, and Bryan Hooi. 2024. Can llms express their uncertainty? an empirical evaluation of confidence elicitation in llms. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Miao+Xiong%2C+Zhiyuan+Hu%2C+Xinyang+Lu%2C+YIFEI+LI%2C+Jie+Fu%2C+Junxian+He%2C+and+Bryan+Hooi%2E+2024%2E+Can+llms+express+their+uncertainty%3F+an+empirical+evaluation+of+confidence+elicitation+in+llms%2E+In+The+Twelfth+In)
39. Tianyang Xu, Shujin Wu, Shizhe Diao, Xiaoze Liu, Xingyao Wang, Yangyi Chen, and Jing Gao. 2024. Sayself: Teaching llms to express confidence with self-reflective rationales. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 5985–5998. [find](https://scholar.google.com/scholar?q=Tianyang+Xu%2C+Shujin+Wu%2C+Shizhe+Diao%2C+Xiaoze+Liu%2C+Xingyao+Wang%2C+Yangyi+Chen%2C+and+Jing+Gao%2E+2024%2E+Sayself%3A+Teaching+llms+to+express+confidence+with+self%2Dreflective+rationales%2E+In+Proceedings+of+the+2024)
40. Xiang Yue, Tianyu Zheng, Yuansheng Ni, Yubo Wang, Kai Zhang, Shengbang Tong, Yuxuan Sun, Botao Yu, Ge Zhang, Huan Sun, and 1 others. 2025. [Mmmu-pro: A more robust multi-discipline multimodal understanding benchmark](https://md.research.elamur.ai/free/en/papers/yue-2024-mmmu-pro-more-robust-multi/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Lingu…
41. Peng Zhang, Wanggui He, Mushui Liu, Wenyi Xiao, Siyu Zou, Yuan Li, Xingjian Wang, Guanghao Zhang, Yanpeng Liu, Weilong Dai, and 1 others. 2026. Fuse: Fine-grained and semantic-aware learning for unified image understanding and generation. In Proceedings of the AAAI Conference on Artificial Intellige… [find](https://scholar.google.com/scholar?q=Peng+Zhang%2C+Wanggui+He%2C+Mushui+Liu%2C+Wenyi+Xiao%2C+Siyu+Zou%2C+Yuan+Li%2C+Xingjian+Wang%2C+Guanghao+Zhang%2C+Yanpeng+Liu%2C+Weilong+Dai%2C+and+1+others%2E+2026%2E+Fuse%3A+Fine%2Dgrained+and+semantic%2Daware+learning+for+unifi)
42. Renrui Zhang, Dongzhi Jiang, Yichi Zhang, Haokun Lin, Ziyu Guo, Pengshuo Qiu, Aojun Zhou, Pan Lu, Kai-Wei Chang, Yu Qiao, Peng Gao, and Hongsheng Li. 2025. [Mathverse: Does your multi-modal llm truly see the diagrams in visual math problems?](https://md.research.elamur.ai/free/en/papers/zhang-2024-mathverse-multi-modal-llm-truly/) In Computer Vision – ECCV 2024, pages 169–186, Cham. Spring…
43. Ruiyang Zhang, Hu Zhang, and Zhedong Zheng. 2024. Vl-uncertainty: Detecting hallucination in large vision-language model via uncertainty estimation. arXiv preprint arXiv:2411.11919. — [arXiv:2411.11919](https://arxiv.org/abs/2411.11919) [find](https://scholar.google.com/scholar?q=Ruiyang+Zhang%2C+Hu+Zhang%2C+and+Zhedong+Zheng%2E+2024%2E+Vl%2Duncertainty%3A+Detecting+hallucination+in+large+vision%2Dlanguage+model+via+uncertainty+estimation%2E+arXiv+preprint+arXiv%3A2411%2E11919%2E+%E2%80%94+%5BarXiv%3A2411%2E1191)
44. Yang Zhao, Kai Xiong, Xiao Ding, Li Du, Yan-gouOuyang, Zhouhao Sun, Jiannan Guan, Wenbin Zhang, Bin Liu, Dong Hu, Bing Qin, and Ting Liu. 2025. UFO-RL: Uncertainty-focused optimization for efficient reinforcement learning data selection. In The Thirty-ninth Annual Conference on Neural Information Pr… [find](https://scholar.google.com/scholar?q=Yang+Zhao%2C+Kai+Xiong%2C+Xiao+Ding%2C+Li+Du%2C+Yan%2DgouOuyang%2C+Zhouhao+Sun%2C+Jiannan+Guan%2C+Wenbin+Zhang%2C+Bin+Liu%2C+Dong+Hu%2C+Bing+Qin%2C+and+Ting+Liu%2E+2025%2E+UFO%2DRL%3A+Uncertainty%2Dfocused+optimization+for+efficient+r)
45. Ziang Zhou, Tianyuan Jin, Jieming Shi, and Li Qing. 2025. Steerconf: Steering LLMs for confidence elicitation. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Ziang+Zhou%2C+Tianyuan+Jin%2C+Jieming+Shi%2C+and+Li+Qing%2E+2025%2E+Steerconf%3A+Steering+LLMs+for+confidence+elicitation%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Information+Processing+Systems%2E)
46. Chengke Zou, Xingang Guo, Rui Yang, Junyu Zhang, Bin Hu, and Huan Zhang. 2024. Dynamath: A dynamic visual benchmark for evaluating mathematical reasoning robustness of vision language models. [find](https://scholar.google.com/scholar?q=Chengke+Zou%2C+Xingang+Guo%2C+Rui+Yang%2C+Junyu+Zhang%2C+Bin+Hu%2C+and+Huan+Zhang%2E+2024%2E+Dynamath%3A+A+dynamic+visual+benchmark+for+evaluating+mathematical+reasoning+robustness+of+vision+language+models%2E)
47. This Appendix for "VL-Calibration: Decoupled Verbalized Confidence for Large Vision-Language Models Reasoning" is organized as follows: • Experimental Setup and Reproducibility. In §A, we describe the experimental setup in detail, including training details (§A.1) and evaluation details (§A.2). The… [find](https://scholar.google.com/scholar?q=This+Appendix+for+%22VL%2DCalibration%3A+Decoupled+Verbalized+Confidence+for+Large+Vision%2DLanguage+Models+Reasoning%22+is+organized+as+follows%3A+%E2%80%A2+Experimental+Setup+and+Reproducibility%2E+In+%C2%A7A%2C+we+describe+the)

*Updated 2 November 2026*