opencompass/TextEdit
TextEdit: A High-Quality, Multi-Scenario Text Editing Benchmark for Generation Models Danni Yang, Sitao Chen, Changyao Tian If you find our work helpful, please give us a ⭐ or cite our paper. See the InternVL-U technical report appendix for more details. 🎉 News [2026/03/06] TextEdit benchmark released. [2026/03/06] Evaluation code and initial baselines released. [2026/03/06] Leaderboard updated with latest models. 📖… See the full description on the dataset page: https://huggingface.co/datasets/opencompass/TextEdit.
Update README.md
Update README.md
Update README.md
Update LICENSE
Upload LICENSE
Update README.md
Update eval_pipeline/vlm_metrics_eval_step2.py
Update eval_pipeline/vlm_metrics_eval_step1.py
Add dataset images via multi-threading API
Update metadata's image url for dataset card viewer
Add metadata.jsonl and dataset card README with image viewer columns
Upload python and shell and prompts about evaluation
initial commit