YANAMI推进多轮多模态有状态交互
AI 总结
论文推出 YANAMI-400K,包含40万段多轮多模态对话、逾200万张图片和260万轮对话,覆盖多图融合、图像编辑、世界建模与视觉推理。配套的 YANAMI-Bench 收录400段多轮对话,用于评估对话状态理解与视觉实现。作者报告称,在 YANAMI-400K 上微调 BAGEL 后,图像编辑、生成和有状态交互分别提升10.3%、2.0%和13.23%。
深度见解
YANAMI 提供了聚焦跨轮维护和更新多模态对话状态的训练与评测数据,论文称现有统一多模态模型在这方面仍面临明显困难。
Computer Science > Computer Vision and Pattern Recognition
arXiv:2611.05280 (cs)
[Submitted on 3 Nov 2026]
Title:YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction
Authors:Shengyu Wang, Tengyue Nong, Junming Liu, Haosong Peng, Siyu Li, Ruiqi Lei, Fushuo Huo, Yuhao Pan, Yunfeng Fan, Ding Wang, Nan Cheng, Haozhao Wang, Wenchao Xu
View a PDF of the paper titled YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction, by Shengyu Wang and 12 other authors