Research map / Vision-language and multimodal / Human and embodied modeling
JourneyDB: A Benchmark for Generative Image Understanding
Keqiang Sun, Junting Pan, Yuying Ge, Hao Li, Haodong Duan, Xiaoshi Wu, Renrui Zhang, Aojun Zhou, Zipeng Qin, Yi Wang, Jifeng Dai, Yu Qiao and 2 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune on JourneyDB for 20 epochs | Uni-Perceiver v2 · JourneyDB prompt inversion test | CIDEr: 15.45 -> 34.01 | Improved |
| fine-tune on JourneyDB | Uni-Perceiver v2 · JourneyDB image captioning val | CIDEr: 21.88 -> 50.72 | Improved |
| fine-tune on JourneyDB | Uni-Perceiver v2 · JourneyDB image captioning test | CIDEr: 20.13 -> 31.76 | Mixed |
| fine-tune on JourneyDB | Uni-Perceiver v2 · JourneyDB prompt inversion test | Similarity: 0.34 -> 0.51 | Improved |
| restrict style retrieval to per-category space instead of overall | CLIP-ViT-L/14 · JourneyDB style retrieval test | recall: 0.47 -> 41.33 | Improved |
| evaluate on style questions instead of content questions | BLIP-2 FlanT5 · JourneyDB MC-VQA test | accuracy (%): 69.7 -> 57.4 | Hurt |
| evaluate generated images instead of real COCO images | BLIP-2 OPT · captioning CIDEr | CIDEr: 145.8 -> 37.60 | Hurt |
Closest papers by meaning
- JourneyBench: A Challenging One-Stop Vision-Language Understanding Benchmark of Generated Images (NeurIPS 2024 · similarity 0.91)
- WEAVE: Unleashing and Benchmarking the In-context Interleaved Comprehension and Generation (CVPR 2026 · similarity 0.89)
- Progressive Compositionality in Text-to-Image Generative Models (ICLR 2025 · similarity 0.88)
- Image Regeneration: Evaluating Text-to-Image Model via Generating Identical Image with Multimodal Large Language Models (AAAI 2025 · similarity 0.88)
- MMIE: Massive Multimodal Interleaved Comprehension Benchmark for Large Vision-Language Models (ICLR 2025 · similarity 0.88)
- OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
