Upload MedAssist-LLM best model (epoch_100) with evaluation results

README.md

@@ -0,0 +1,98 @@
+---
+license: apache-2.0
+library_name: transformers
+---
+# MedAssist-LLM
+<!-- markdownlint-disable first-line-h1 -->
+<!-- markdownlint-disable html -->
+<!-- markdownlint-disable no-duplicate-header -->
+
+<div align="center">
+  <img src="figures/fig1.png" width="60%" alt="MedAssist-LLM" />
+</div>
+<hr>
+
+<div align="center" style="line-height: 1;">
+  <a href="LICENSE" style="margin: 2px;">
+    <img alt="License" src="figures/fig2.png" style="display: inline-block; vertical-align: middle;"/>
+  </a>
+</div>
+
+## 1. Introduction
+
+MedAssist-LLM is a specialized healthcare language model designed for clinical decision support and medical knowledge assistance. This model has been fine-tuned on extensive medical literature, clinical notes, and healthcare datasets to provide accurate and reliable medical information assistance.
+
+<p align="center">
+  <img width="80%" src="figures/fig3.png">
+</p>
+
+The model excels at understanding complex medical terminology, interpreting clinical scenarios, and providing evidence-based recommendations. It has been evaluated across multiple healthcare benchmarks including clinical diagnosis, drug interactions, and medical question answering.
+
+**Important**: This model is designed to assist healthcare professionals and should not replace professional medical judgment. Always consult qualified healthcare providers for medical decisions.
+
+## 2. Evaluation Results
+
+### Comprehensive Medical Benchmark Results
+
+<div align="center">
+
+| | Benchmark | BioMedLM | ClinicalBERT | MedPaLM | MedAssist-LLM |
+|---|---|---|---|---|---|
+| **Clinical Tasks** | Clinical Diagnosis | 0.623 | 0.651 | 0.712 | 0.660 |
+| | Drug Interaction | 0.589 | 0.612 | 0.678 | 0.671 |
+| | Adverse Event Detection | 0.534 | 0.567 | 0.623 | 0.696 |
+| **Knowledge Tasks** | Medical Q&A | 0.701 | 0.723 | 0.789 | 0.750 |
+| | Medical Literature Comprehension | 0.645 | 0.678 | 0.734 | 0.675 |
+| | EHR Understanding | 0.567 | 0.598 | 0.656 | 0.679 |
+| **Diagnostic Tasks** | Radiology Analysis | 0.512 | 0.545 | 0.612 | 0.581 |
+| | Pathology Classification | 0.578 | 0.601 | 0.667 | 0.713 |
+| | Patient Risk Assessment | 0.523 | 0.556 | 0.623 | 0.711 |
+| **Clinical Support** | Treatment Recommendation | 0.612 | 0.645 | 0.701 | 0.744 |
+| | Medical Coding | 0.489 | 0.523 | 0.589 | 0.570 |
+| | Clinical Notes Summarization | 0.634 | 0.667 | 0.723 | 0.761 |
+
+</div>
+
+### Overall Performance Summary
+MedAssist-LLM demonstrates strong performance across all medical benchmark categories, with particularly notable results in clinical decision support and medical knowledge tasks.
+
+## 3. Clinical API Access
+We offer a secure API for healthcare institutions to integrate MedAssist-LLM. Please contact our team for enterprise access and HIPAA-compliant deployment options.
+
+## 4. How to Run Locally
+
+Please refer to our code repository for detailed instructions on running MedAssist-LLM in your clinical environment.
+
+### Security Considerations
+1. This model processes sensitive medical data. Ensure proper data handling procedures.
+2. Deploy in HIPAA-compliant infrastructure when processing PHI.
+3. Implement appropriate access controls and audit logging.
+
+### System Prompt
+We recommend using the following system prompt:
+```
+You are MedAssist-LLM, a medical AI assistant designed to support healthcare professionals.
+Today is {current date}.
+Important: Always recommend consulting qualified healthcare providers for medical decisions.
+```
+
+### Temperature
+We recommend setting the temperature parameter to 0.3 for medical applications to ensure consistent and reliable outputs.
+
+### Prompts for Clinical Documents
+For clinical document analysis:
+```
+clinical_template = \
+"""[Document Type]: {doc_type}
+[Patient Context]: {patient_context}
+[Clinical Content Begin]
+{clinical_content}
+[Clinical Content End]
+{query}"""
+```
+
+## 5. License
+This code repository is licensed under the [Apache 2.0 License](LICENSE). Use of MedAssist-LLM in clinical settings requires compliance with applicable healthcare regulations.
+
+## 6. Contact
+For questions or enterprise inquiries, please contact us at support@medassist-llm.health.

config.json

@@ -0,0 +1,10 @@
+{
+  "model_type": "llama",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "hidden_size": 4096,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "vocab_size": 32000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2704b4ff968d3c6124109066db38f189838ed72317d63c76298a2a8a8bacb85a
+size 74