ResearchNov 4, 2026, 00:00 UTC

YANAMI dataset and benchmark target multi-turn stateful multimodal interaction

YANAMI-400K contains 400K multi-turn multimodal dialogues with over 2M images, and YANAMI-Bench has 400 dialogues. Fine-tuning BAGEL on it improved image editing by 10.3%, generation by 2.0% and stateful interaction by 13.23%.

ModelBAGEL
ModelYANAMI-400K
ModelYANAMI-Bench
Topicresearch
Topicbenchmark
Topicimage

1outlet ran itRead at Hugging Face Papers

Status: Research
Published: Nov 4, 2026, 00:00 UTC That day
Outlets: 1
Written by: Claude How

Every outlet that ran it 1 article

Nov 4, 2026, 00:00 UTC
Hugging Face PapersResearchLead
YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction

All news