{"data":{"slug":"ukplab-gpl","name":"gpl","tagline":"Unsupervised domain adaptation method for dense retrieval using generative pseudo labeling","github_url":"https://github.com/UKPLab/gpl","owner":"UKPLab","repo":"gpl","owner_avatar_url":"https://avatars.githubusercontent.com/u/9532046?v=4","primary_language":"Python","stars":342,"forks":38,"topics":["bert","domain-adaptation","information-retrieval","nlp","transformers","vector-search"],"archived":false,"github_pushed_at":"2023-07-06T09:58:34+00:00","maintenance_label":"Dormant","stars_delta_30d":-1,"url":"https://www.graphcanon.com/tools/ukplab-gpl","markdown_url":"https://www.graphcanon.com/tools/ukplab-gpl.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/ukplab-gpl","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=ukplab-gpl","description":"Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled corpus and yields massive improvement: \"GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval\" https://arxiv.org/abs/2112.07577","homepage_url":null,"license":"Apache-2.0","open_issues":26,"watchers":4,"ai_summary":"GPL technique to adapt dense retrieval models through unlabeled data, enhancing performance in new domains without labeled data.","readme_excerpt":"## Installation\nOne can either install GPL via `pip`\n```bash\npip install gpl\n```\nor via `git clone`\n```bash\ngit clone https://github.com/UKPLab/gpl.git && cd gpl\npip install -e .\n```\n> Meanwhile, please make sure the [correct version of PyTorch](https://pytorch.org/get-started/locally/) has been installed according to your CUDA version.","github_created_at":"2021-12-14T16:14:03+00:00","created_at":"2026-07-11T11:36:08.82221+00:00","updated_at":"2026-09-20T06:26:01.280267+00:00","categories":[{"slug":"data-retrieval","name":"Data & Retrieval","url":"https://www.graphcanon.com/categories/data-retrieval","markdown_url":"https://www.graphcanon.com/categories/data-retrieval.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/data-retrieval"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"bert","name":"bert"},{"slug":"domain-adaptation","name":"domain-adaptation"},{"slug":"information-retrieval","name":"information retrieval"},{"slug":"nlp","name":"nlp"},{"slug":"transformers","name":"transformers"},{"slug":"vector-search","name":"vector-search"}],"trust":{"provenance":{"is_fork":false,"github_id":438319137,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T06:25:58.435Z","maintenance":{"label":"Dormant","score":18,"methodology":"github_public_v1","releases_90d":0,"days_since_push":1171,"last_release_at":"2022-09-29T15:09:58Z","stars_delta_30d":-1,"open_issues_delta_30d":0},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:36:10.103Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-20T06:25:59.706Z"},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-09-20T06:25:59.706Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-20T06:25:59.706Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you have an abundance of unlabeled data from a target domain but lack labeled data.","For improving information retrieval systems where labeling is costly or time-consuming."],"when_not_to_use":["Avoid when high precision and recall on labeled datasets are critical in the initial phase without adaptation.","If significant computational resources for unsupervised learning are not available, then GPL may not be suitable."],"source":"enrich:decision_facts","observed_at":"2026-07-16T20:38:01.440Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"GPL enhances dense retrieval models by adapting them to new domains without the need for labeled data, relying solely on unlabeled corpora."}]}}