Research map / Vision-language and multimodal / Cross-modal learning
T-APT: Text-Guided Modality-Aware Prompt Tuning for Arbitrary Multimodal Remote Sensing Data Joint Classification
Qinghao Gao, Jiahui Qu, Wenqian Dong
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 reconstruction loss Lrec (ablation) | T-APT · MUUFL (LiDAR+HS) | OA (%): 89.21 -> 90.90 | Improved |
| add alignment loss Lalign (ablation) | T-APT · Houston2013 (MS+HS) | OA (%): 97.95 -> 98.60 | Improved |
| add both Lrec and Lalign (ablation) | T-APT · Augsburg (HS+SAR) | OA (%): 90.21 -> 91.80 | Improved |
| use prompt tuning (MAC) instead of full fine-tuning (ablation) | T-APT · Augsburg (HS+SAR) | OA (%): 92.38 -> 91.80 | Hurt |
| use prompt tuning (MAC) instead of full fine-tuning (ablation) | T-APT · Houston2013 (MS+HS) | OA (%): 98.25 -> 98.60 | Improved |
| add text-guided prompt (ablation) | T-APT · Augsburg (HS+SAR+LiDAR) | OA (%): 91.84 -> 93.17 | Improved |
| add text-guided prompt (ablation) | T-APT · Augsburg (HS+SAR) | OA (%): 91.99 -> 91.80 | No clear effect |
| add text-guided prompt (ablation) | T-APT · Houston2013 (HSI+LiDAR) | OA (%): 98.36 -> 98.89 | Improved |
| keep specific-shared interleaved Mamba in MAMF-FC instead of replacing it (ablation) | T-APT · Houston2013, MUUFL, Augsburg | OA (%) | Improved |
| use MAC prompt tuning instead of adapter-only fine-tuning (ablation) | T-APT · Houston2013, MUUFL, Augsburg | OA (%): degradations of 1.32%, 1.43%, and 2.76% | Improved |
| fuse more modalities (add LiDAR to HS+SAR) | T-APT · Augsburg | OA (%): 91.80 -> 93.17 | Improved |
Closest papers by meaning
- CF-IPT: Cross-Modal Fusion Interactive Prompt Tuning of Vision-Language Pre-Trained Model for Multisource Remote Sensing Data Classification (CVPR 2026 · similarity 0.91)
- SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery (CVPR 2024 · similarity 0.87)
- LDS2AE: Local Diffusion Shared-Specific Autoencoder for Multimodal Remote Sensing Image Classification with Arbitrary Missing Modalities (AAAI 2024 · similarity 0.85)
- RobSense: A Robust Multi-modal Foundation Model for Remote Sensing with Static, Temporal, and Incomplete Data Adaptability (CVPR 2025 · similarity 0.85)
- MMM-RS: A Multi-modal, Multi-GSD, Multi-scene Remote Sensing Dataset and Benchmark for Text-to-Image Generation (NeurIPS 2024 · similarity 0.85)
- Galileo: Learning Global & Local Features of Many Remote Sensing Modalities (ICML 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.
