Research map / Vision-language and multimodal / Cross-modal learning
StructXLIP: Enhancing Vision-language Models with Multimodal Structural Cues
Zanxi Ruan, Songqun Gao, Qiuyu Kong, Yiming Wang, Marco Cristani
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 |
|---|---|---|---|
| add structure-centric losses (global edge-text, consistency, local) to the base contrastive loss (ablation) | StructXLIP · SKETCHY | R@1 T->I: 66.75 (global loss only) -> 69.86 (all three) | Improved |
| add consistency regularization loss (ablation) | StructXLIP · SKETCHY | R@1 T->I: 66.75 -> 67.88 | Improved |
| add local structure-centric loss (ablation) | StructXLIP · DOCCI | R@1 T->I: 81.92 -> 83.04 | Improved |
| add L* structure losses to FineLIP | FineLIP · SKETCHY | R@1 T->I: 40.59 -> 59.15 | Improved |
| add L* structure losses to SmartCLIP | SmartCLIP · SKETCHY | R@1 T->I: 50.73 -> 52.94 | Improved |
| add L* structure losses to GOAL | GOAL · SKETCHY | R@1 T->I: 63.21 -> 67.88 | Improved |
| add L* structure losses to LoRA | LoRA · SKETCHY | R@1 T->I: 57.08 -> 62.09 | Improved |
| add L* structure losses to SigLIP2 | SigLIP2 · SKETCHY | R@1 T->I: 68.91 -> 73.49 | Improved |
| use lexicon filter instead of original captions (ablation) | StructXLIP · SKETCHY | R@1 T->I: 66.93 -> 69.86 | Improved |
| use lexicon filter instead of LLM rewriting (ablation) | StructXLIP · DOCCI | R@1 T->I: 81.08 -> 83.04 | Improved |
| use LoG instead of Canny edges (ablation) | StructXLIP · DOCCI | R@1 T->I: 83.04 -> 83.59 | No clear effect |
| use learned edge extractors (HED, LAD, P2S) instead of Canny (ablation) | StructXLIP · SKETCHY | R@1 T->I: 69.86 -> 68.14 to 69.17 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- SmartCLIP — SmartCLIP: Modular Vision-language Alignment with Identification Guarantees (CVPR 2025) · better on 3 of 3 shared results
- Long-CLIP — Long-CLIP: Unlocking the Long-Text Capability of CLIP (ECCV 2024) · better on 3 of 3 shared results
- FineLIP — FineLIP: Extending CLIP's Reach via Fine-Grained Alignment with Longer Text Inputs (CVPR 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FineLIP: Extending CLIP's Reach via Fine-Grained Alignment with Longer Text Inputs (CVPR 2025 · similarity 0.91)
- Unlocking Fine-Grained Perception in CLIP via Structurally-Aware Latent Masked Modeling (NeurIPS 2026 · similarity 0.91)
- SmartCLIP: Modular Vision-language Alignment with Identification Guarantees (CVPR 2025 · similarity 0.91)
- Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding (NeurIPS 2026 · similarity 0.90)
- Linguistic-Aware Patch Slimming Framework for Fine-grained Cross-Modal Alignment (CVPR 2024 · similarity 0.90)
- AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Document Understanding (NeurIPS 2025 · similarity 0.89)
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.
