MuSLR: Multimodal Symbolic Logical Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917229569769472 |
|---|---|
| author | Xu, Jundong Fei, Hao Zhang, Yuhui Pan, Liangming Huang, Qijun Liu, Qian Nakov, Preslav Kan, Min-Yen Wang, William Yang Lee, Mong-Li Hsu, Wynne |
| author_facet | Xu, Jundong Fei, Hao Zhang, Yuhui Pan, Liangming Huang, Qijun Liu, Qian Nakov, Preslav Kan, Min-Yen Wang, William Yang Lee, Mong-Li Hsu, Wynne |
| contents | Multimodal symbolic logical reasoning, which aims to deduce new facts from multimodal input via formal logic, is critical in high-stakes applications such as autonomous driving and medical diagnosis, as its rigorous, deterministic reasoning helps prevent serious consequences. To evaluate such capabilities of current state-of-the-art vision language models (VLMs), we introduce the first benchmark MuSLR for multimodal symbolic logical reasoning grounded in formal logical rules. MuSLR comprises 1,093 instances across 7 domains, including 35 atomic symbolic logic and 976 logical combinations, with reasoning depths ranging from 2 to 9. We evaluate 7 state-of-the-art VLMs on MuSLR and find that they all struggle with multimodal symbolic reasoning, with the best model, GPT-4.1, achieving only 46.8%. Thus, we propose LogiCAM, a modular framework that applies formal logical rules to multimodal inputs, boosting GPT-4.1's Chain-of-Thought performance by 14.13%, and delivering even larger gains on complex logics such as first-order logic. We also conduct a comprehensive error analysis, showing that around 70% of failures stem from logical misalignment between modalities, offering key insights to guide future improvements. All data and code are publicly available at https://llm-symbol.github.io/MuSLR. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25851 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MuSLR: Multimodal Symbolic Logical Reasoning Xu, Jundong Fei, Hao Zhang, Yuhui Pan, Liangming Huang, Qijun Liu, Qian Nakov, Preslav Kan, Min-Yen Wang, William Yang Lee, Mong-Li Hsu, Wynne Computer Vision and Pattern Recognition Multimodal symbolic logical reasoning, which aims to deduce new facts from multimodal input via formal logic, is critical in high-stakes applications such as autonomous driving and medical diagnosis, as its rigorous, deterministic reasoning helps prevent serious consequences. To evaluate such capabilities of current state-of-the-art vision language models (VLMs), we introduce the first benchmark MuSLR for multimodal symbolic logical reasoning grounded in formal logical rules. MuSLR comprises 1,093 instances across 7 domains, including 35 atomic symbolic logic and 976 logical combinations, with reasoning depths ranging from 2 to 9. We evaluate 7 state-of-the-art VLMs on MuSLR and find that they all struggle with multimodal symbolic reasoning, with the best model, GPT-4.1, achieving only 46.8%. Thus, we propose LogiCAM, a modular framework that applies formal logical rules to multimodal inputs, boosting GPT-4.1's Chain-of-Thought performance by 14.13%, and delivering even larger gains on complex logics such as first-order logic. We also conduct a comprehensive error analysis, showing that around 70% of failures stem from logical misalignment between modalities, offering key insights to guide future improvements. All data and code are publicly available at https://llm-symbol.github.io/MuSLR. |
| title | MuSLR: Multimodal Symbolic Logical Reasoning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.25851 |