Research map / Vision-language and multimodal / CLIP and zero-shot
BiomedCCPL: Causal Conditional Prompt Learning for Biomedical Vision-Language Models
Xueliang Cui, Juncai Zhang, Jiacheng Hou, Dan Lu, Hao Zhang, Ruxin Wang
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 SCD to the CoOp baseline (ablation) | CoOp baseline · Base-to-Novel, avg 11 datasets | Novel accuracy (%): 64.72 -> 75.69 | Improved |
| add VGAP to the CoOp baseline (ablation) | CoOp baseline · Base-to-Novel, avg 11 datasets | Novel accuracy (%): 64.72 -> 73.26 | Improved |
| add SAR to the CoOp baseline (ablation) | CoOp baseline · Base-to-Novel, avg 11 datasets | Novel accuracy (%): 64.72 -> 65.68 | Improved |
| add VGAP to SAR + SCD (ablation) | BiomedCCPL · Few-shot, avg 11 datasets, 16-shot | accuracy (%): 71.13 -> 82.25 | Improved |
| add SCD to SAR + VGAP (ablation) | BiomedCCPL · Few-shot, avg 11 datasets, 16-shot | accuracy (%): 79.80 -> 82.25 | Improved |
| add SAR to VGAP + SCD (ablation) | BiomedCCPL · Few-shot, avg 11 datasets, 16-shot | accuracy (%): 77.48 -> 82.25 | Improved |
| add SCD to SAR + VGAP (ablation) | BiomedCCPL · Base-to-Novel, avg 11 datasets | Novel accuracy (%): 73.87 -> 79.20 | Improved |
| add SAR to VGAP alone for few-shot (ablation) | VGAP only · Few-shot, avg 11 datasets, 16-shot | accuracy (%): 79.95 -> 79.80 | No clear effect |
| add adaptive prototypes in VGAP (ablation) | BiomedCCPL · Base-to-Novel, avg 11 datasets | HM (%): 76.91 -> 79.98 | Improved |
| add adaptive prototypes in VGAP (ablation) | BiomedCCPL · Few-shot, avg 11 datasets, 16-shot | accuracy (%): 78.20 -> 82.25 | Improved |
| use BiomedCCPL instead of BiomedCoOp on novel classes | Base-to-Novel · Base-to-Novel, avg 10 datasets | Novel accuracy (%): 70.46 -> 79.20 | Improved |
| use BiomedCCPL instead of MMA on harmonic mean | Base-to-Novel · Base-to-Novel, avg 10 datasets | HM (%): 73.53 -> 79.98 | Improved |
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.
- MMRL — MMRL: Multi-Modal Representation Learning for Vision-Language Models (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
- BiomedCoOp: Learning to Prompt for Biomedical Vision-Language Models (CVPR 2025 · similarity 0.92)
- BioDPP: Dynamic Prompt Policy Learning for Biomedical Vision-Language Models (AAAI 2026 · similarity 0.91)
- LLaDA-MedV: Exploring Large Language Diffusion Models for Biomedical Image Understanding (CVPR 2026 · similarity 0.88)
- MedUnifier: Unifying Vision-and-Language Pre-training on Medical Data with Vision Generation Task using Discrete Visual Representations (CVPR 2025 · similarity 0.87)
- Prompt as Knowledge Bank: Boost Vision-language model via Structural Representation for zero-shot medical detection (ICLR 2025 · similarity 0.87)
- ExGra-Med: Extended Context Graph Alignment for Medical Vision-Language Models (NeurIPS 2025 · similarity 0.87)
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.
