CharXiv: Charting Gaps in Realistic Chart Understanding in …
2024年6月26日 · Our results reveal a substantial, previously underestimated gap between the reasoning skills of the strongest proprietary model (i.e., GPT-4o), which achieves 47.1% …
GitHub - princeton-nlp/CharXiv: [NeurIPS 2024] CharXiv: Charting Gaps ...
To ensure quality, all charts and questions are handpicked, curated, and verified by human experts. Our results reveal a substantial, previously underestimated gap between the …
CharXiv: Charting Gaps in Realistic Chart Understanding in...
2024年9月26日 · TL;DR: CharXiv reveals significant shortcomings in MLLMs' chart understanding, showing a large performance gap between models and humans. Chart …
CharXiv
Human experts curated and verified all charts and questions. Our findings show a significant gap in reasoning skills, with the strongest proprietary model (GPT-4o) achieving 47.1% accuracy …
陈丹琦团队发布CharXiv数据集：重新定义图表理解的评估标准-CSD…
2024年6月28日 · 这篇文章的标题是《CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs》，由普林斯顿大学、威斯康星大学麦迪逊分校和香港大学的研究人员撰写。 …
CharXiv: Charting Gaps in Realistic Chart Understanding in …
Our results reveal a substantial, previously underestimated gap between the reasoning skills of the strongest proprietary model (i.e., GPT-4o), which achieves 47.1% accuracy, and the …
【数据集】最近开源的一些多模态图表理解数据集 - 文章 - 开发者 …
为了解决这个问题，作者提出了一个新的评估套件CharXiv，它包含了从arXiv论文中精选的2323个自然、具有挑战性和多样性的图表，并设计了两种类型的问题：描述性问题和推理问题，以全 …
CharXiv Dataset - Papers With Code
The results from CharXiv reveal a substantial gap between the reasoning skills of the strongest proprietary model (i.e., GPT-4o), which achieves 47.1% accuracy, and the strongest open …
【数据集】最近开源的一些多模态图表理解数据集_charxiv-CSDN …
2024年7月3日 · 为了解决这个问题，作者提出了一个新的评估套件CharXiv，它包含了从 arXiv 论文中精选的2323个自然、具有挑战性和多样性的图表，并设计了两种类型的问题：描述性问 …
CharXiv: Charting Gaps in Realistic Chart Understanding in …
2024年6月26日 · 该论文揭示了最强专有模型（GPT-4o）和最强开源模型（InternVL Chat V1.5）之间推理能力的实质性差距，最强专有模型的准确率为47.1％，而最强开源模型的准确 …
- 某些结果已被删除