Research map / Vision-language and multimodal / CLIP and zero-shot
TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias
Sanghyun Jo, Soohyun Ryu, Sungyub Kim, Eunho Yang, Kyungsu Kim
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 L_distill (text-tag self-distillation) (ablation) | TCL fine-tuning · Pascal VOC / text-level | mIoU: 58.5 -> 61.1 | Improved |
| add L_tag only (without L_distill) (ablation) | TCL fine-tuning · Text-level segmentation | CaptionIoU: 60.4 -> 60.7 | No clear effect |
| select tags by pixel-tag scoring instead of NLTK (ablation) | TTD pseudo-tagging · Pascal VOC | mIoU: 56.5 -> 61.1 | Improved |
| select tags by global image-tag scoring instead of pixel-tag scoring (ablation) | TTD pseudo-tagging · Pascal VOC | mIoU: 55.0 -> 52.5 | Hurt |
| add TTD fine-tuning to MaskCLIP | MaskCLIP · 6 open-vocab segmentation benchmarks (avg) | mIoU: 19.0 -> 27.5 | Improved |
| add TTD fine-tuning to TCL | TCL · 6 open-vocab segmentation benchmarks (avg) | mIoU: 30.4 -> 33.9 | Improved |
| use pixel-tag scoring for multi-tag selection instead of global image scoring | TCL encoders · CC3M val multi-tag selection | F1: 43.7 -> 78.5 | Improved |
| prune training samples to those with above mean+std image-text similarity (ablation) | TTD training · Pascal VOC | mIoU: 60.5 -> 61.1 with 247 -> 34 hours | Improved |
| apply TTD versus TagAlign on large-class datasets | TCL + TTD · COCO-Stuff, ADE20k | mIoU: COCO-Stuff 23.7 vs 25.3; ADE 17.0 vs 17.3 | 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.
- MaskCLIP — MaskCLIP: Masked Self-Distillation Advances Contrastive Language-Image Pretraining (CVPR 2023) · 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
- Improving Long-Text Alignment for Text-to-Image Diffusion Models (ICLR 2025 · similarity 0.87)
- Mining Fine-Grained Image-Text Alignment for Zero-Shot Captioning via Text-Only Training (AAAI 2024 · similarity 0.86)
- CLIP in Mirror: Disentangling text from visual images through reflection (NeurIPS 2024 · similarity 0.86)
- Discriminative Probing and Tuning for Text-to-Image Generation (CVPR 2024 · similarity 0.86)
- Seeing What Matters: Empowering CLIP with Patch Generation-to-Selection (CVPR 2025 · similarity 0.86)
- Retaining Knowledge and Enhancing Long-Text Representations in CLIP through Dual-Teacher Distillation (CVPR 2025 · similarity 0.86)
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.
