{"data":{"slug":"zorazrw-filco","name":"filco","tagline":"Learning to Filter Context for Retrieval-Augmented Generation","github_url":"https://github.com/zorazrw/filco","owner":"zorazrw","repo":"filco","owner_avatar_url":"https://avatars.githubusercontent.com/u/49503887?v=4","primary_language":"Python","stars":199,"forks":20,"topics":["dialog-generation","fact-verification","flan-t5","llama2","question-answering","retrieval-augmented-generation"],"archived":false,"github_pushed_at":"2024-04-06T18:57:53+00:00","maintenance_label":"Dormant","stars_delta_30d":1,"url":"https://www.graphcanon.com/tools/zorazrw-filco","markdown_url":"https://www.graphcanon.com/tools/zorazrw-filco.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/zorazrw-filco","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=zorazrw-filco","description":"[Preprint] Learning to Filter Context for Retrieval-Augmented Generaton","homepage_url":"https://arxiv.org/pdf/2311.08377.pdf","license":"CC-BY-SA-4.0","open_issues":2,"watchers":1,"ai_summary":"A Python-based project focusing on dialog generation, fact verification, and question answering using models like Flan-T5 and LLaMA2. It employs Dense Passage Retriever for retrieving relevant Wikipedia passages.","readme_excerpt":"## Install\n\nInstall all required libraries by running\n\n```bash\npip install -r requirements.txt\n```\n\nRetrieve top relevant Wikipedia passages using [Dense Passage Retriever (DPR)](https://github.com/facebookresearch/DPR)\nand store into the `./datasets/${name}` directory. We also provide preprocessed datasets with top-5 retrieved passages [(here)](https://drive.google.com/file/d/13z_qrVOBlgu75IJBpX-1vMSCC6hC9yH4/view?usp=sharing).\nWe specify `${name}` for six datasets with ['nq', 'tqa', 'hotpotqa', 'fever', 'wow'] in following example commands.","github_created_at":"2023-11-14T21:18:30+00:00","created_at":"2026-07-11T23:07:28.194186+00:00","updated_at":"2026-09-20T04:29:09.109926+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":"dialog-generation","name":"dialog-generation"},{"slug":"fact-verification","name":"fact-verification"},{"slug":"flan-t5","name":"flan-t5"},{"slug":"llama2","name":"llama2"},{"slug":"question-answering","name":"question-answering"},{"slug":"retrieval-augmented-generation","name":"retrieval-augmented generation"}],"trust":{"provenance":{"is_fork":false,"github_id":718814616,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:13:18.683Z","maintenance":{"label":"Dormant","score":18,"methodology":"github_public_v1","releases_90d":0,"days_since_push":894,"last_release_at":null,"stars_delta_30d":1,"open_issues_delta_30d":0},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":32,"high_count":7,"last_scan_at":"2026-09-18T08:13:20.937Z","medium_count":12,"scan_profile":"deps","critical_count":1}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:13:17.521Z"},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-09-18T08:13:17.521Z"},"license_spdx":{"value":"CC-BY-SA-4.0","source":"github.license","observed_at":"2026-09-18T08:13:17.521Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need to generate dialogues that require fact verification and question answering capabilities, leveraging models like Flan-T5 and LLaMA2.","If your project involves retrieval-augmented generation and you want to use preprocessed datasets with top-5 retrieved passages from Wikipedia.","For tasks that require the integration of Dense Passage Retriever for retrieving relevant passages from Wikipedia."],"when_not_to_use":["If your project does not require the use of Flan-T5 or LLaMA2 models for dialog generation, fact verification, or question answering.","When you do not need to retrieve relevant Wikipedia passages using Dense Passage Retriever.","If your project does not involve the use of preprocessed datasets with top-5 retrieved passages from Wikipedia."],"source":"enrich:decision_facts","observed_at":"2026-09-18T09:10:01.201Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Filco is a Python-based project that focuses on dialog generation, fact verification, and question answering using models like Flan-T5 and LLaMA2, with Dense Passage Retriever for retrieving relevant Wikipedia passages."}]}}