Elamur Research⌕
☰
home / papers / 2026 / SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models

SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models

Paper · 2026 · Weijiang Lv, Yaoxuan Feng, Xiaobo Xia et al.

Summary

Chain-of-Thought traces from multimodal large language models (MLLMs) often look coherent, yet they may not reflect how visual evidence actually drives the answer. Earlier work mostly measured perceptual hallucinations. This paper targets reasoning-level unfaithfulness instead. The authors built SPD-Faith Bench on fine-grained image difference reasoning. Comparing two nearly identical pictures forces models to use visual evidence and blocks language-prior shortcuts. Testing 12 open-source and proprietary MLLMs revealed two recurring failures: perceptual blindness and perception-reasoning dissociation. Mechanistic analysis linked both to visual attention that fades during reasoning and to representation shifts in the residual stream. Based on these findings, the authors propose SAGE, a training-free framework that strengthens visual routing and aligns generated reasoning with what the model perceives. SAGE improves faithfulness metrics on the new benchmark and also reduces hallucinations on MME and CHAIR. The study shows that faithfulness needs its own evaluation, separate from answer correctness.

Key points

  • Benchmark design. SPD-Faith Bench has 3,000 image pairs in two subsets:
    • a single-difference subset, split into easy, medium and hard by instance density;
    • a multi-difference subset with 2–5 controlled changes.
  • Edit construction. Each edit is a color change, an object removal or a position change. GPT-4o planned the edits, LaMa inpainting carried them out, and humans verified the results.
  • Evaluation protocol. Metrics cover three dimensions:
    • overall perception of the image pair, scored with DS and DQR;
    • how faithfully differences are perceived, scored with TF1 and CF1;
    • how faithfully the model reasons, scored with CR and DRF.
  • SAGE on SPD-Faith Bench. With Qwen2.5-VL-7B, SAGE raised CR from 38.2 to 43.8 and DRF from 29.5 to 33.7 over greedy decoding, beating VCD, SC, SR, API and Zoom-Refine.
  • SAGE on other benchmarks and limitations.
    • On MME, SAGE scored 1702.81, compared with 1649.13 for AGLA.
    • On CHAIR with LLaVA-1.5-7B, it reached C_S 42.3, C_I 12.7 and Recall 81.8.
    • The evaluation is limited to image difference captioning.

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

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 works

🧭 Topics 3Research subjects and languages the work is about.Language models & pretraining · 3,320 worksReasoning · 1,426 worksVision & language · 657 works

Cite

Cite as Weijiang Lv, Yaoxuan Feng, Xiaobo Xia, Jiayu Wang, Yan Jing, Wenchao Chen and Bo Chen. 2026. SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.995

@inproceedings{lv2026spd,
  title = {{SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models}},
  author = {Weijiang Lv, Yaoxuan Feng, Xiaobo Xia, Jiayu Wang, Yan Jing, Wenchao Chen and Bo Chen},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.995},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.995}
}

References

75 references, 33 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Wenbin An, Feng Tian, Sicong Leng, Jiahao Nie, Haonan Lin, QianYing Wang, Ping Chen, Xiaoqin Zhang, and Shijian Lu. 2025. Mitigating object hallucinations in large vision-language models with assembly of global and local attention. In CVPR, pages 29915– 29926. find
  2. Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2016. Spice: Semantic propositional image caption evaluation. In ECCV, pages 382–398.
  3. Iván Arcuschin, Jett Janiak, Robert Krzyzanowski, Senthooran Rajamanoharan, Neel Nanda, and Arthur Conmy. 2025. Chain-of-thought reasoning in the wild is not always faithful. arXiv preprint arXiv:2503.08679. find
  4. Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. 2023. Qwen-vl: A frontier large vision-language model with versatile abilities. arXiv preprint arXiv:2308.12966, 1(2):3.
  5. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923. — Qwen2.5-VL Technical Report
  6. Zechen Bai, Pichao Wang, Tianjun Xiao, Tong He, Zongbo Han, Zheng Zhang, and Mike Zheng Shou. 2024. Hallucination of multimodal large language models: A survey. arXiv preprint arXiv:2404.18930.
  7. Satanjeev Banerjee and Alon Lavie. 2005. Meteor: An automatic metric for mt evaluation with improved correlation with human judgments. In ACL Workshop, pages 65–72.
  8. Fazl Barez, Tung-Yu Wu, Iván Arcuschin, Michael Lan, Vincent Wang, Noah Siegel, Nicolas Collignon, Clement Neo, Isabelle Lee, Alasdair Paren, and 1 others. 2025. Chain-of-thought is not explainability. Preprint, alphaXiv, page v1. find
  9. Alexander Black, Jing Shi, Yifei Fan, Tu Bui, and John Collomosse. 2024. Vixen: Visual text comparison network for image difference captioning. In AAAI, pages 846–854. find
  10. Lin Chen, Jinsong Li, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Zehui Chen, Haodong Duan, Jiaqi Wang, Yu Qiao, Dahua Lin, and 1 others. 2024a. Are we on the right way for evaluating large vision-language models? In NeurIPS, pages 27056–27087.
  11. Zhe Chen, Weiyun Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Erfei Cui, Jinguo Zhu, Shenglong Ye, Hao Tian, Zhaoyang Liu, and 1 others. 2024b. Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling. arXiv preprint arXiv:2412.05271.
  12. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, and 1 others. 2025. Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities. arX…
  13. Wenliang Dai, Junnan Li, Dongxu Li, Anthony Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale N Fung, and Steven Hoi. 2023. Instructblip: Towards general-purpose vision-language models with instruction tuning. In NeurIPS, pages 49250–49267.
  14. Yexing Du, Ziyang Ma, Yifan Yang, Keqi Deng, Xie Chen, Bo Yang, Yang Xiang, Ming Liu, and Bing Qin. 2024. Cot-st: Enhancing llm-based speech translation with multimodal chain-of-thought. arXiv preprint arXiv:2409.19510. — [arXiv:2409.19510](https://arxiv.org/abs/2409.19510) find
  15. Jesseba Fernando and Grigori Guitchounts. 2025. Transformer dynamics: A neuroscientific approach to interpretability of large language models. arXiv preprint arXiv:2502.12131. — [arXiv:2502.12131](https://arxiv.org/abs/2502.12131) find
  16. Maxwell Forbes, Christine Kaeser-Chen, Piyush Sharma, and Serge Belongie. 2019. Neural naturalist: Generating fine-grained image comparisons. arXiv preprint arXiv:1909.04101. find
  17. Chaoyou Fu, Peixian Chen, Yunhang Shen, Yulei Qin, Mengdan Zhang, Xu Lin, Jinrui Yang, Xiawu Zheng, Ke Li, Xing Sun, and 1 others. 2025. Mme: A comprehensive evaluation benchmark for multimodal large language models. In NeurIPS Datasets and Benchmarks Track.
  18. Timin Gao, Peixian Chen, Mengdan Zhang, Chaoyou Fu, Yunhang Shen, Yan Zhang, Shengchuan Zhang, Xiawu Zheng, Xing Sun, Liujuan Cao, and 1 others. 2024. Cantor: Inspiring multimodal chain-of-thought of mllm. In ACM MM, pages 9096–9105. find
  19. Mor Geva, Avi Caciularu, Kevin Wang, and Yoav Goldberg. 2022. Transformer feed-forward layers build predictions by promoting concepts in the vocabulary space. In EMNLP, pages 30–45. others. 2024. Gpt-4o system card. arXiv preprint arXiv:2410.21276.
  20. Harsh Jhamtani and Taylor Berg-Kirkpatrick. 2018. Learning to describe differences between pairs of similar images. arXiv preprint arXiv:1808.10584. find
  21. Jingjing Jiang, Chao Ma, Xurui Song, Hanwang Zhang, and Jun Luo. 2025. Corvid: Improving multimodal large language models towards chain-of-thought reasoning. In ICCV, pages 3034–3046. find
  22. Tamera Lanham, Anna Chen, Ansh Radhakrishnan, Benoit Steiner, Carson Denison, Danny Hernandez, Dustin Li, Esin Durmus, Evan Hubinger, Jackson Kernion, and 1 others. 2023. Measuring faithfulness in chain-of-thought reasoning. arXiv preprint arXiv:2307.13702.
  23. Sicong Leng, Hang Zhang, Guanzheng Chen, Xin Li, Shijian Lu, Chunyan Miao, and Lidong Bing. 2024. Mitigating object hallucinations in large vision-language models through visual contrastive decoding. In CVPR, pages 13872–13882.
  24. Junxian Li, Xinyue Xu, Sai Ma, and Sichao Li. 2025a. Faithact: Faithfulness planning and acting in mllms. arXiv preprint arXiv:2511.08409. — [arXiv:2511.08409](https://arxiv.org/abs/2511.08409) find
  25. Yi Li, Hualiang Wang, Xinpeng Ding, Haonan Wang, and Xiaomeng Li. 2025b. Token activation map to visually explain multimodal llms. arXiv preprint arXiv:2506.23270. — [arXiv:2506.23270](https://arxiv.org/abs/2506.23270) find
  26. Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Wayne Xin Zhao, and Ji-Rong Wen. 2023. Evaluating object hallucination in large vision-language models. arXiv preprint arXiv:2305.10355.
  27. Chin-Yew Lin. 2004. Rouge: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81.
  28. Mor Geva, Roei Schuster, Jonathan Berant, and Omer Levy. 2021. Transformer feed-forward layers are key-value memories. In EMNLP, pages 5484–5495. find
  29. 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. In ECCV, pages 740– 755.
  30. Meng-Hao Guo, Jiajun Xu, Yi Zhang, Jiaxi Song, Haoyang Peng, Yi-Xuan Deng, Xinzhi Dong, Kiyohiro Nakayama, Zhengyang Geng, Chen Wang, and 1 others. 2025. R-bench: Graduate-level multidisciplinary benchmarks for llm & mllm complex reasoning evaluation. In ICML. find
  31. Zihao Lin, Samyadeep Basu, Mohammad Beigi, Varun Manjunatha, Ryan A Rossi, Zichao Wang, Yufan Zhou, Sriram Balasubramanian, Arman Zarei, Keivan Rezaei, and 1 others. 2025. A survey on mechanistic interpretability for multi-modal foundation models. arXiv preprint arXiv:2502.17516. — [arXiv:2502.17516](https://arxiv.org/abs/2502.17516) find
  32. 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 CVPR, pages 13418–13427. find
  33. Shi Liu, Kecheng Zheng, and Wei Chen. 2024a. Paying more attention to image: A training-free method for alleviating hallucination in lvlms. In ECCV, pages 125–140. find
  34. Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, and 1 Yuan Liu, Haodong Duan, Yuanhan Zhang, Bo Li, Songyang Zhang, Wangbo Zhao, Yike Yuan, Jiaqi Wang, Conghui He, Ziwei Liu, and 1 others. 2024b. Mmbench: Is yo…
  35. Zhining Liu, Ziyi Chen, Hui Liu, Chen Luo, Xianfeng Tang, Suhang Wang, Joy Zeng, Zhenwei Dai, Zhan Shi, Tianxin Wei, and 1 others. 2025a. Seeing but not believing: Probing the disconnect between visual attention and answer correctness in vlms. arXiv preprint arXiv:2510.17771. — [arXiv:2510.17771](https://arxiv.org/abs/2510.17771) find
  36. Hao Shao, Shengju Qian, Han Xiao, Guanglu Song, Zhuofan Zong, Letian Wang, Yu Liu, and Hongsheng Li. 2024. Visual cot: Advancing multimodal language models with a comprehensive dataset and benchmark for chain-of-thought reasoning. In NeurIPS, pages 8612–8642. find
  37. Zujing Liu, Junwen Pan, Qi She, Yuan Gao, and Guisong Xia. 2025b. On the faithfulness of visual thinking: Measurement and enhancement. arXiv preprint arXiv:2510.23482. — [arXiv:2510.23482](https://arxiv.org/abs/2510.23482) find
  38. Xu Shen, Song Wang, Zhen Tan, Laura Yao, Xinyu Zhao, Kaidi Xu, Xin Wang, and Tianlong Chen. 2025. Faithcot-bench: Benchmarking instance-level faithfulness of chain-of-thought reasoning. arXiv preprint arXiv:2510.04040. — [arXiv:2510.04040](https://arxiv.org/abs/2510.04040) find
  39. Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, and Jianfeng Gao. 2023. Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts. arXiv preprint arXiv:2310.02255.
  40. Run Luo, Lu Wang, Wanwei He, Longze Chen, Jiaming Li, and Xiaobo Xia. 2025a. Gui-r1: A generalist r1-style vision-language action model for gui agents. arXiv preprint arXiv:2504.10458. — [arXiv:2504.10458](https://arxiv.org/abs/2504.10458) find
  41. Run Luo, Xiaobo Xia, Lu Wang, Longze Chen, Renke Shan, Jing Luo, Min Yang, and Tat-Seng Chua. 2026. Next-omni: Towards any-to-any omnimodal foundation models with discrete flow matching. In ICLR. find
  42. Run Luo, Haonan Zhang, Longze Chen, Ting-En Lin, Xiong Liu, Yuchuan Wu, Min Yang, Yongbin Li, Minzheng Wang, Pengpeng Zeng, and 1 others. 2025b. Mmevol: Empowering multimodal large language models with evol-instruct. In ACL Findings, pages 19655–19682. find
  43. Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, and 1 others. 2023. Self-refine: Iterative refinement with self-feedback. In NeurIPS, pages 46534–46594. find
  44. Chancharik Mitra, Brandon Huang, Trevor Darrell, and Roei Herzig. 2024. Compositional chain-of-thought prompting for large multimodal models. In CVPR, pages 14420–14431. find
  45. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In ACL, pages 311–318.
  46. Dong Huk Park, Trevor Darrell, and Anna Rohrbach. 2019. Robust change captioning. In ICCV, pages 4624–4633. find
  47. Yi Peng, Peiyu Wang, Xiaokun Wang, Yichen Wei, Jiangbo Pei, Weijie Qiu, Ai Jian, Yunzhuo Hao, Jiachun Pan, Tianyidan Xie, and 1 others. 2025. Skywork r1v: Pioneering multimodal reasoning with chain-of-thought. arXiv preprint arXiv:2504.05599.
  48. Anna Rohrbach, Lisa Anne Hendricks, Kaylee Burns, Trevor Darrell, and Kate Saenko. 2018. Object hallucination in image captioning. arXiv preprint arXiv:1809.02156.
  49. Jiajie Su, Haoyuan Wang, Xiaohua Feng, Yunshan Ma, Xiaobo Xia, Yuyuan Li, Xiaolin Zheng, Jianmao Xiao, and Chaochao Chen. 2026. Out-of-distribution generalization via invariant trajectories for multimodal large language model editing. arXiv preprint arXiv:2601.19700. — [arXiv:2601.19700](https://arxiv.org/abs/2601.19700) find
  50. Roman Suvorov, Elizaveta Logacheva, Anton Mashikhin, Anastasia Remizova, Arsenii Ashukha, Aleksei Silvestrov, Naejin Kong, Harshith Goka, Kiwoong Park, and Victor Lempitsky. 2021. Resolution-robust large mask inpainting with fourier convolutions. arXiv preprint arXiv:2109.07161. — [arXiv:2109.07161](https://arxiv.org/abs/2109.07161) find
  51. Anand Swaroop, Akshat Nallani, Saksham Uboweja, Adiliia Uzdenova, Michael Nguyen, Kevin Zhu, Sunishchal Dev, Ashwinee Panda, Vasu Sharma, and Maheep Chaudhary. 2025. Frit: Using causal importance to improve chain-of-thought faithfulness. arXiv preprint arXiv:2509.13334. — [arXiv:2509.13334](https://arxiv.org/abs/2509.13334) find
  52. Hao Tan, Franck Dernoncourt, Zhe Lin, Trung Bui, and Mohit Bansal. 2019. Expressing visual relationships via language. arXiv preprint arXiv:1906.07689. find
  53. V Team, Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, Shuaiqi Duan, Weihan Wang, Yan Wang, Yean Cheng, Zehai He, Zhe Su, Zhen Yang, Ziyang Pan, and 69 others. 2025. Glm-4.5v and glm-4.1v-thinking: Towards versatile multimodal reas…
  54. Miles Turpin, Julian Michael, Ethan Perez, and Samuel Bowman. 2023. Language models don’t always say what they think: Unfaithful explanations in chain-of-thought prompting. In NeurIPS, pages 74952–74965.
  55. Ramakrishna Vedantam, C Lawrence Zitnick, and Devi Parikh. 2015. Cider: Consensus-based image description evaluation. In CVPR, pages 4566–4575.
  56. Haifeng Wang, Jiwei Li, Hua Wu, Eduard Hovy, and Yu Sun. 2023. Pre-trained language models and their applications. Engineering, 25:51–65. find
  57. Ke Wang, Junting Pan, Weikang Shi, Zimu Lu, Houxing Ren, Aojun Zhou, Mingjie Zhan, and Hongsheng Li. 2024a. Measuring multimodal mathematical reasoning with math-vision dataset. In NeurIPS, pages 95095–95169.
  58. Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, and 1 others. 2024b. Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution. arXiv preprint arXiv:2409.12191.
  59. Xintong Wang, Jingheng Pan, Liang Ding, and Chris Biemann. 2024c. Mitigating hallucinations in large vision-language models with instruction contrastive decoding. arXiv preprint arXiv:2403.18715. find
  60. 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. arXiv preprint arXiv:2203.11171.
  61. Yaoting Wang, Shengqiong Wu, Yuecheng Zhang, Shuicheng Yan, Ziwei Liu, Jiebo Luo, and Hao Fei. 2025. Multimodal chain-of-thought reasoning: A comprehensive survey. arXiv preprint arXiv:2503.12605. — [arXiv:2503.12605](https://arxiv.org/abs/2503.12605) find
  62. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. Chain-of-thought prompting elicits reasoning in large language models. In NeurIPS, pages 24824–24837.
  63. Lai Wei, Wenkai Wang, Xiaoyu Shen, Yu Xie, Zhihao Fan, Xiaojin Zhang, Zhongyu Wei, and Wei Chen. 2024. Mc-cot: A modular collaborative cot framework for zero-shot medical-vqa with llm and mllm integration. arXiv preprint arXiv:2410.04521. — [arXiv:2410.04521](https://arxiv.org/abs/2410.04521) find
  64. Qiong Wu, Xiangcong Yang, Yiyi Zhou, Chenxin Fang, Baiyang Song, Xiaoshuai Sun, and Rongrong Ji. 2025. Grounded chain-of-thought for multimodal large language models. arXiv preprint arXiv:2503.12799. — [arXiv:2503.12799](https://arxiv.org/abs/2503.12799) find
  65. Xuan Yu, Dayan Guan, and Yanfeng Gu. 2025. Zoom-refine: Boosting high-resolution multimodal understanding via localized zoom and self-refinement. arXiv preprint arXiv:2506.01663. — [arXiv:2506.01663](https://arxiv.org/abs/2506.01663) find
  66. Zeping Yu and Sophia Ananiadou. 2024. Neuron-level knowledge attribution in large language models. In EMNLP, pages 3267–3280. find
  67. Daoan Zhang, Junming Yang, Hanjia Lyu, Zijian Jin, Yuan Yao, Mingkai Chen, and Jiebo Luo. 2024a. Cocot: Contrastive chain-of-thought prompting for large multimodal models with multiple image inputs. arXiv preprint arXiv:2401.02582. — [arXiv:2401.02582](https://arxiv.org/abs/2401.02582) find
  68. Duzhen Zhang, Yahan Yu, Jiahua Dong, Chenxing Li, Dan Su, Chenhui Chu, and Dong Yu. 2024b. Mm-llms: Recent advances in multimodal large language models. arXiv preprint arXiv:2401.13601.
  69. Jiarui Zhang, Mahyar Khayatkhoei, Prateek Chhikara, and Filip Ilievski. 2025a. Mllms know where to look: Training-free perception of small visual details with multimodal llms. arXiv preprint arXiv:2502.17422. — [arXiv:2502.17422](https://arxiv.org/abs/2502.17422) find
  70. Jusheng Zhang, Kaitong Cai, Xiaoyang Guo, Sidi Liu, Qinhan Lv, Ruiqi Chen, Jing Yang, Yijia Fan, Xiaofei Sun, Jian Wang, and 1 others. 2025b. Mm-cot: A benchmark for probing visual chain-of-thought reasoning in multimodal models. arXiv preprint arXiv:2512.08228. — [arXiv:2512.08228](https://arxiv.org/abs/2512.08228) find
  71. Zhuosheng Zhang, Aston Zhang, Mu Li, Hai Zhao, George Karypis, and Alex Smola. 2023. Multimodal chain-of-thought reasoning in language models. arXiv preprint arXiv:2302.00923.
  72. Zhiyu Wu, Xiaokang Chen, Zizheng Pan, Xingchao Liu, Wen Liu, Damai Dai, Huazuo Gao, Yiyang Ma, Chengyue Wu, Bingxuan Wang, and 1 others. 2024. Deepseek-vl2: Mixture-of-experts vision-language models for advanced multimodal understanding. arXiv preprint arXiv:2412.10302.
  73. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  74. Yuan Yao, Tianyu Yu, Ao Zhang, Chongyi Wang, Junbo Cui, Hongji Zhu, Tianchi Cai, Haoyu Li, Weilin Zhao, Zhihui He, and 1 others. 2024. Minicpm-v: A gpt-4v level mllm on your phone. arXiv preprint arXiv:2408.01800.
  75. Runpeng Yu, Weihao Yu, and Xinchao Wang. 2024. Attention prompting on image for large vision-language models. In ECCV, pages 251–268. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/lv-2026-spd-faith-bench-diagnosing-improving/