YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction
cs.CV
Submitted: 2026-11-03
Updated: 2026-11-03
Project page: https://rowletqwq.github.io/YANAMI
Terminology
Sources
- BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset
- EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing
- Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
- MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data
- Emu3.5: Native Multimodal Models are World Learners
- ChatUMM: Robust Context Tracking for Conversational Interleaved Generation
- Emerging Properties in Unified Multimodal Pretraining
- GoT: Unleashing Reasoning Capability of Multimodal Large Language Model for Visual Generation and Editing
- FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
- Zebra-CoT: A Dataset for Interleaved Vision Language Reasoning
- Step1X-Edit: A Practical Framework for General Image Editing
- ConvBench: A Multi-Turn Conversation Evaluation Benchmark with Hierarchical Capability for Large Vision-Language Models
- Transfer between Modalities with MetaQueries
- Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification
- Chameleon: Mixed-Modal Early-Fusion Foundation Models
- InternVL-U: Democratizing Unified Multimodal Models for Understanding, Reasoning, Generation and Editing
- Ovis-U1 Technical Report
- Emu3: Next-Token Prediction is All You Need
- GPT-IMAGE-EDIT-1.5M: A Million-Scale, GPT-Generated Image Dataset
- Skywork UniPic 3.0: Unified Multi-Image Composition via Sequence Modeling
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs