Explore and submit NER models
Browse and submit evaluation results for AI benchmarks
Simulate causal effects and determine variable control
Evaluate diversity in data sets to improve fairness
Browse and compare Indic language LLMs on a leaderboard
Explore speech recognition model performance
Check system health
Search and save datasets generated with a LLM in real time
Analyze and visualize car data
This project is a GUI for the gpustack/gguf-parser-go
Generate plots for GP and PFN posterior approximations
Filter and view AI model leaderboard data
Transfer GitHub repositories to Hugging Face Spaces
The Clinical NER Leaderboard is a platform designed to evaluate and compare Named Entity Recognition (NER) models specifically tailored for clinical and medical text data. It provides a centralized hub for researchers and developers to submit their models, benchmark performance, and explore state-of-the-art solutions in clinical NLP.
• Model Comparison: Allows users to compare performance metrics of different NER models on clinical datasets.
• Benchmark Scores: Provides standardized benchmark scores for clinical NER tasks, enabling apples-to-apples comparisons.
• Interactive Visualization: Offers dynamic visualizations to explore model performance across different entity types and datasets.
• Model Submission: Enables researchers to submit their own NER models for evaluation and inclusion in the leaderboard.
• Community Engagement: Facilitates discussion and collaboration through forums and shared resources.
What is Named Entity Recognition (NER) in the clinical context?
NER in clinical contexts involves identifying and categorizing entities such as diseases, medications, symptoms, and genes from unstructured clinical text.
How can I submit my NER model to the leaderboard?
Visit the platform's submission page, follow the provided guidelines, and upload your model along with required documentation.
What datasets are used for benchmarking on the leaderboard?
The leaderboard uses standardized clinical datasets, including publicly available sources like MIMIC and i2b2, to ensure fair and consistent evaluations.