Research map / Vision-language and multimodal / CLIP and zero-shot
Long-CLIP: Unlocking the Long-Text Capability of CLIP
Beichen Zhang, Pan Zhang, Xiaoyi Dong, Yuhang Zang, Jiaqi 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 knowledge-preserved stretching of positional embedding instead of fixed-ratio interpolation (ablation) | Long-CLIP · ImageNet zero-shot | top-1 accuracy: 58.8 -> 66.8 | Improved |
| add primary component matching (ablation) | Long-CLIP with KPS · CIFAR-100 zero-shot | top-1 accuracy: 65.9 -> 69.3 | Improved |
| add primary component matching (ablation) | Long-CLIP with KPS · Urban-200 text-to-image | R@1: 78.0 -> 79.0 | Improved |
| add KPS without PCM versus neither (ablation) | Long-CLIP · COCO text-to-image | R@5: 43.4 -> 64.3 | Improved |
| naive fixed-ratio interpolation and long-text fine-tuning versus original CLIP | CLIP ViT-B/16 · ImageNet zero-shot | top-1 accuracy: 68.4 -> 55.1 | Hurt |
| use PCA-based primary component matching versus undistinguished image feature (ablation) | Long-CLIP · Urban-200 text-to-image | R@1: 71.0 -> 79.0 | Improved |
| use PCM instead of mixed-length text (ablation) | Long-CLIP · Urban-200 text-to-image | R@1: 67.5 -> 79.0 | Improved |
| use PCM instead of bounded text encoder (SmoothL1) (ablation) | Long-CLIP · CIFAR-100 zero-shot | top-1 accuracy: 67.9 -> 69.3 | Improved |
| Long-CLIP versus CLIP on zero-shot classification | CLIP ViT-L/14 · ImageNet zero-shot | top-1 accuracy: 75.5 -> 73.5 | Hurt |
| Long-CLIP replaces CLIP on short-caption retrieval | CLIP ViT-B/16 · COCO 5k text-to-image | R@1: 32.7 -> 40.4 | Improved |
| apply KPS and PCM to DeCLIP ViT-B/32 | DeCLIP · Urban long-caption retrieval | T2I/I2T R@1: 28.0/25.5 -> 54.0/51.0 | Improved |
Later papers that compared with this method
17 later accepted papers report Long-CLIP as a baseline in their results tables; 16 of them report a better number on most of the results they share with it.
- CLIP Is Shortsighted: Paying Attention Beyond the First Sentence (CVPR 2026 · better on 6 of 6 shared results)
- OneLIP: Unlocking and Improving Long-Text Representations of CLIP via One-Stage Adaptation (AAAI 2026 · better on 6 of 6 shared results)
- HiMo-CLIP: Modeling Semantic Hierarchy and Monotonicity in Vision-Language Alignment (AAAI 2026 · better on 5 of 5 shared results)
- Role-SynthCLIP: A Role-Play Driven Diverse Synthetic Data Approach (CVPR 2026 · better on 3 of 4 shared results)
- StructXLIP: Enhancing Vision-language Models with Multimodal Structural Cues (CVPR 2026 · better on 3 of 3 shared results)
- b-CLIP: Text-Conditioned Contrastive Learning for Multi-Granular Vision-Language Alignment (CVPR 2026 · better on 3 of 3 shared results)
- DialogueVPR: Towards Conversational Visual Place Recognition (CVPR 2026 · better on 2 of 2 shared results)
- ReasonCLIP-58M: Visually Grounded Commonsense Reasoning Supervision for CLIP (ECCV 2026 · better on 1 of 2 shared results)
9 more not listed.
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
- Retaining Knowledge and Enhancing Long-Text Representations in CLIP through Dual-Teacher Distillation (CVPR 2025 · similarity 0.93)
- Demystifying CLIP Data (ICLR 2024 · similarity 0.91)
- FG-CLIP: Fine-Grained Visual and Textual Alignment (ICML 2025 · similarity 0.91)
- Meta CLIP 2: A Worldwide Scaling Recipe (NeurIPS 2025 · similarity 0.90)
- LoTLIP: Improving Language-Image Pre-training for Long Text Understanding (NeurIPS 2024 · similarity 0.90)
- FineLIP: Extending CLIP's Reach via Fine-Grained Alignment with Longer Text Inputs (CVPR 2025 · similarity 0.90)
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.
