{"id":23885433,"url":"https://github.com/shalinianandaphd/prism","last_synced_at":"2026-06-17T04:31:23.650Z","repository":{"id":270165155,"uuid":"909198010","full_name":"ShaliniAnandaPhD/PRISM","owner":"ShaliniAnandaPhD","description":"pRISM is a repository that combines Retrieval-Augmented Generation (RAG) with a multi-LLM voting approach to create accurate and reliable AI-generated outputs. It integrates multiple language models, including Mistral, Claude 3.5, and OpenAI, to enhance performance through advanced consensus techniques","archived":false,"fork":false,"pushed_at":"2025-06-20T03:23:14.000Z","size":15072,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-20T04:18:56.904Z","etag":null,"topics":["ai-transparency","document-processing-pipeline","fine-tuning","legal-ai","llm","lora","multi-model","phi-4","semantic-ai","weighted-majority"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ShaliniAnandaPhD.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":".github/FUNDING.yml","license":"LICENSE.md","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},"funding":{"github":"ShaliniAnandaPhD"}},"created_at":"2024-12-28T02:04:28.000Z","updated_at":"2025-06-20T03:23:17.000Z","dependencies_parsed_at":"2024-12-29T00:24:25.459Z","dependency_job_id":null,"html_url":"https://github.com/ShaliniAnandaPhD/PRISM","commit_stats":null,"previous_names":["shalinianandaphd/prism"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ShaliniAnandaPhD/PRISM","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShaliniAnandaPhD%2FPRISM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShaliniAnandaPhD%2FPRISM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShaliniAnandaPhD%2FPRISM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShaliniAnandaPhD%2FPRISM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ShaliniAnandaPhD","download_url":"https://codeload.github.com/ShaliniAnandaPhD/PRISM/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShaliniAnandaPhD%2FPRISM/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34434492,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-17T02:00:05.408Z","response_time":127,"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":["ai-transparency","document-processing-pipeline","fine-tuning","legal-ai","llm","lora","multi-model","phi-4","semantic-ai","weighted-majority"],"created_at":"2025-01-04T04:49:41.476Z","updated_at":"2026-06-17T04:31:23.646Z","avatar_url":"https://github.com/ShaliniAnandaPhD.png","language":"Python","funding_links":["https://github.com/sponsors/ShaliniAnandaPhD"],"categories":[],"sub_categories":[],"readme":"Open Source Attribution Notice\n\nUsing open source software without proper attribution or in violation of license terms is not only ethically problematic but may also constitute a legal violation. I believe in supporting the open source community that makes projects like this possible.\n\nIf you're using code or tools from this repository or GitHub, please ensure you maintain all attribution notices and comply with all applicable licenses.\n\nThe license above is a modified MIT LICENSE for the purpose of this project 👆\n\n(The project is ongoing and updates will be made going into 2025 )\n\n# pRISM\n*Procedural Intelligence \u0026 RAG-based Semantic Multi-Model*\n\npRISM combines RAG (Retrieval-Augmented Generation) with multi-model consensus to create more reliable AI outputs, specifically designed for legal document processing. By integrating multiple language models and incorporating human expert feedback, we aim to make AI-generated legal content more transparent and trustworthy.\n\n\u003e **Note**: This is a research project - please don't use it for actual legal work without expert review!\n\n## What Makes pRISM Different?\n\n- **Multi-Model Consensus**: Instead of relying on a single LLM, we aggregate outputs from multiple models using techniques like weighted majority voting. In our testing, this has significantly reduced hallucination rates (full benchmarks coming soon).\n\n- **Domain-Specific Fine-Tuning**: We use LoRA to adapt models for legal tasks without breaking the bank on compute costs. Our initial tests show promising improvements in legal terminology accuracy.\n\n- **Human-in-the-Loop**: Legal experts review outputs and provide feedback, which gets incorporated back into the system to improve future results.\n\n## Demos\n\n- [CLI Tool Demo for Prism](https://www.linkedin.com/posts/shalinianandaphd_cli-demo-for-prism-continuation-from-my-activity-7305376243787382784-rdm3)\n- [Evaluations](https://www.linkedin.com/posts/shalinianandaphd_legaltech-accesstojustice-ai-activity-7305289571565518848-xSOX)\n- [Prism: Proactive Retrieval Intelligence - Architecture](https://www.linkedin.com/posts/shalinianandaphd_prism-proactive-retrieval-intelligence-activity-7303428785339543552-UIzc)\n\n## Quick Start\n\n1. Clone and install:\n```bash\ngit clone https://github.com/your-repo/pRISM.git\ncd pRISM\npip install -r requirements.txt\n```\n\n## Core Components\n\n### RAG Pipeline (`rag_pipeline.py`)\nThe heart of pRISM. Handles document retrieval and generation using FAISS for vector search and LangChain for orchestration. \n\n### Real-Time API (`real_time_inference_api.py`)\nFastAPI service for running inferences. Currently supports:\n- Basic document queries\n- Multi-model consensus outputs\n- Confidence scoring\n- Explanation generation\n\n### Evaluation Tools\n- `performance_benchmarking.py`: Measures latency and accuracy\n- `evaluation_report_generator.py`: Generates detailed PDF reports\n- `automated_feedback_loops.py`: Incorporates user feedback for continuous improvement\n\n\n\n## Contributing\nThis is a research project and we'd love your input! Feel free to open issues or PRs, especially if you have experience with:\n- Legal document processing\n- RAG implementations\n- Model fine-tuning\n- Evaluation methodologies\n\n## License\nMIT License - see LICENSE.md\n\n---\nBuilt with 💡 for exploring better ways to handle legal AI\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshalinianandaphd%2Fprism","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fshalinianandaphd%2Fprism","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshalinianandaphd%2Fprism/lists"}