{"id":51521264,"url":"https://osu-nlp-group.github.io/GUI-Drag/","last_synced_at":"2026-07-27T14:00:32.160Z","repository":{"id":318961740,"uuid":"1073905304","full_name":"OSU-NLP-Group/GUI-Drag","owner":"OSU-NLP-Group","description":null,"archived":false,"fork":false,"pushed_at":"2026-01-19T03:19:06.000Z","size":119678,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-01-31T11:55:42.168Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://osu-nlp-group.github.io/GUI-Drag","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/OSU-NLP-Group.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-10T20:03:54.000Z","updated_at":"2026-01-19T03:19:11.000Z","dependencies_parsed_at":"2025-10-17T12:43:16.259Z","dependency_job_id":null,"html_url":"https://github.com/OSU-NLP-Group/GUI-Drag","commit_stats":null,"previous_names":["osu-nlp-group/gui-drag"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/OSU-NLP-Group/GUI-Drag","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OSU-NLP-Group%2FGUI-Drag","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OSU-NLP-Group%2FGUI-Drag/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OSU-NLP-Group%2FGUI-Drag/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OSU-NLP-Group%2FGUI-Drag/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/OSU-NLP-Group","download_url":"https://codeload.github.com/OSU-NLP-Group/GUI-Drag/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OSU-NLP-Group%2FGUI-Drag/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35953520,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-07-20T02:08:10.276Z","status":"online","status_checked_at":"2026-07-27T02:00:06.776Z","response_time":101,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-07-08T16:00:43.736Z","updated_at":"2026-07-27T14:00:32.138Z","avatar_url":"https://github.com/OSU-NLP-Group.png","language":"Python","funding_links":[],"categories":["[Paper List](#paperlist)"],"sub_categories":["GUI Grounding Benchmark"],"readme":"# Beyond Clicking: A Step Towards Generalist GUI Grounding via Text Dragging\n\n\n\u003e Graphical user interface (GUI) grounding, the process of mapping human instructions to GUI actions, serves as a fundamental basis to autonomous GUI agents. While existing grounding models achieve promising performance to simulate the mouse click action on various click-based benchmarks, another essential mode of mouse interaction, namely dragging, remains largely underexplored. Yet, dragging the mouse to select and manipulate textual content represents a prevalent and important usage in practical GUI scenarios. To narrow this gap, we first introduce \\textsc{GUI-Drag}, a diverse dataset of 161K text dragging examples synthesized through a scalable pipeline. To support systematic and robust evaluation, we further construct \\textsc{ScreenDrag}, a benchmark with 5,333 examples spanning three levels of interface context, together with three dedicated metrics designed for assessing text dragging capability. Models trained on \\textsc{GUI-Drag} with an efficient continual training strategy achieve substantial improvements on \\textsc{ScreenDrag}, while preserving the original click-based performance on ScreenSpot, ScreenSpot-v2, and OSWorld-G. Our work encourages further research on broader GUI grounding beyond just clicking and paves way toward a truly generalist GUI grounding model.\n\n![GUI-Drag Dataset Overview](assets/GUI_Drag.png)\n\n\n\n## Setup\n\n- Python 3.12\n- `pip install -r requirement.txt`\n\n\n\n## Experiments\n\n1. Unzip the `image_and_ocr.zip` in the root folder, which contains the examples of the ScreenDrag benchmark and OCR results used during evaluation.\n\n2. Follow the commands in `evaluation/cli_run_drag.sh`. It supports the models containing `GUI-Drag-3/7B (via vllm)`, `Claude computer use`, `OpenAI computer use` and `UI-Tars`.\n\n3. Calculate the metrics of the Success Rate and B-Dist via `metrics/cli_run_drag_metric_new.sh`. It will first output the metric results and summarize the model performance in a report with different breakdowns.\n\n\n### Credentials\n\n| Backend | Required environment variables |\n| --- | --- |\n| Claude CUA | `AWS_REGION`, `AWS_ACCESS_KEY`, `AWS_SECRET_KEY` (for Bedrock access) |\n| OpenAI Operator | `OPENAI_API_KEY` |\n\n\n## Datasets and Models\n\n`GUI-Drag` dataset can be downloaded at [here](https://huggingface.co/datasets/osunlp/GUI-Drag-dataset).\n\n`GUI-Drag-3/7B` models, which are trained via efficient continual training, can be accessed at [here](https://huggingface.co/osunlp/GUI-Drag-7B).\n\n## Benchmarks\n\nThe `ScreenDrag` benchmark can be found in the `benchmark.json` file. Note that you should first unzip the `image_and_ocr.zip` file to support the evaluation.\n\n## Citations\n\nIf you find our data, model, benchmark or the general resources useful, please consider citing us via:\n\n```bash\n@article{liao2025beyond,\n  title={Beyond Clicking: A Step Towards Generalist GUI Grounding via Text Dragging},\n  author={Liao, Zeyi and Lu, Yadong and Gou, Boyu and Sun, Huan and Awadallah, Ahmed},\n  journal={arXiv preprint arXiv:2601.06031},\n  year={2025}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/osu-nlp-group.github.io%2FGUI-Drag%2F","html_url":"https://awesome.ecosyste.ms/projects/osu-nlp-group.github.io%2FGUI-Drag%2F","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/osu-nlp-group.github.io%2FGUI-Drag%2F/lists"}