Research map / Detection and segmentation / Open-set and few-shot detection
CrossVL: Complexity-Aware Feature Routing and Paired Curriculum for Cross-View Vision-Language Detection
Zhipeng Liu, Chunbo Luo
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 CPA (ablation) | Florence-2-base · MAVREC aerial test | mAP: 58.66 -> 60.66 | Improved |
| add paired curriculum (ablation) | Florence-2-base · MAVREC aerial test | mAP: 58.66 -> 56.53 | Hurt |
| add paired curriculum (ablation) | Florence-2-base · MAVREC aerial val | mAP: 63.73 -> 64.37 | Improved |
| add CPA and paired curriculum (ablation) | Florence-2-base · MAVREC aerial test | mAP: 58.66 -> 61.03 | Improved |
| add CPA and paired curriculum (ablation) | Florence-2-base · MAVREC aerial val | mAP: 63.73 -> 65.35 | Improved |
| add single linear pathway without routing (ablation) | Florence-2-base · MAVREC aerial test | mAP: 58.66 -> 60.09 | Improved |
| replace single-path with full multi-path CPA (ablation) | CPA · MAVREC aerial test | mAP: 60.09 -> 60.66 | Improved |
| replace single-path with full multi-path CPA (ablation) | CPA · MAVREC aerial test | mAP_M: 68.65 -> 85.69 | Improved |
| replace single-path with full multi-path CPA (ablation) | CPA · MAVREC aerial val | mAP: 64.68 -> 64.49 | No clear effect |
| add CPA and paired curriculum (ablation) | Florence-2-base · MAVREC test ground-aerial gap | mAP gap: 8.63 -> 6.65 | Improved |
| add CPA alone (ablation) | Florence-2-base · MAVREC test ground-aerial gap | mAP gap: 8.63 -> 9.30 | Hurt |
| add paired curriculum alone (ablation) | Florence-2-base · MAVREC test ground-aerial gap | mAP gap: 8.63 -> 12.59 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- YOLOv7 — YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors (CVPR 2023) · better on 1 of 1 shared result
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
- UniBench: Visual Reasoning Requires Rethinking Vision-Language Beyond Scaling (NeurIPS 2024 · similarity 0.86)
- MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs (ICML 2026 · similarity 0.85)
- Synthesize Diagnose and Optimize: Towards Fine-Grained Vision-Language Understanding (CVPR 2024 · similarity 0.85)
- GroundVLP: Harnessing Zero-Shot Visual Grounding from Vision-Language Pre-training and Open-Vocabulary Object Detection (AAAI 2024 · similarity 0.85)
- Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation (ICLR 2026 · similarity 0.85)
- Dynamic Multimodal Evaluation with Flexible Complexity by Vision-Language Bootstrapping (ICLR 2025 · similarity 0.85)
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.
